2024/07/28 更新

お知らせ

 

写真a

  
VASCONCELLOS VARGAS DANILO
VARGAS VASCONCELLOS DANILO
所属
システム情報科学研究院 情報学部門 准教授
システム情報科学研究院 情報学部門(併任)
工学部 電気情報工学科(併任)
システム情報科学府 情報理工学専攻(併任)
職名
准教授
プロフィール
(最近の活動や研究は研究室のサイトにある) 研究に関して、次世代の人工知能を作成するために研究しています。さらに、共同研究をしながら、最先端の人工知能を様々な分野(セキュリティ、電気電子工学、ロボット工学、など)に応用しています。成果はトップ雑誌と学会に公開しています。平成29年にBBCニュースに公開された研究もありました。他の研究は平成29年にニューラルネットワークの分野の中に最もインパクトがある雑誌に公開されました。  研究を活発的に行えるように、助成金に応募しています。平成30年と平成29年には、ACT-I「情報と未来」を含めて、 三つの助成金が採択されました。  教育に関して、新しい科目(人工知能Ⅰ:基礎を理解する と人工知能Ⅱ:最先端を理解する)を開き、トップレベル学会(Core Rank A)にチュートリアルを開き、人工知能のサークルも作成しました。
ホームページ
外部リンク

学位

  • 博士(学術)

研究テーマ・研究キーワード

  • 研究テーマ: 自己組織化による学習:人工知能の新たな基礎を開拓

    研究キーワード: 自己組織化、人工知能

    研究期間: 2021年4月 - 2025年4月

  • 研究テーマ: 頑強な潜在変数の学習に関する研究

    研究キーワード: 敵対的機械学習

    研究期間: 2020年4月 - 2022年4月

  • 研究テーマ: 人間の知覚に基づくロバストと安全な人工知能

    研究キーワード: 敵対的機械学習

    研究期間: 2020年4月 - 2022年4月

  • 研究テーマ: 進化的計算、大域的最適化;

    研究キーワード: 進化的計算

    研究期間: 2017年3月

  • 研究テーマ: アプリケーション:サイバーセキュリティー、ロボット工学, 電気電子工学(例えば, 経済給電)、脳科学、医学 など。

    研究キーワード: サイバーセキュリティー, ロボット工学, 電気電子工学, 脳科学、医学

    研究期間: 2017年3月

  • 研究テーマ: 機会学習、(ディープ)ニューラルネットワーク、Neuroevolution;

    研究キーワード: (ディープ)ニューラルネットワーク

    研究期間: 2017年3月

  • 研究テーマ: 汎用人工知能、認知アーキテクチャ;

    研究キーワード: 人工汎用知能

    研究期間: 2016年11月

受賞

  • IEEE Transactions on Evolutionary Computation Outstanding Paper Award 2022

    2022年7月   IEEE   Outstanding paper

  • 2016 Excellent Student Award of The IEEE Fukuoka Section

    2017年2月   IEEE Fukuoka Section  

論文

  • Continual General Chunking Problem and SyncMap 査読 国際誌

    D. V. Vargas, T. Asabuki

    AAAI 2021   2020年12月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

    Humans possess an inherent ability to chunk sequences into their constituent parts. In fact, this ability is thought to bootstrap language skills and learning of image patterns which might be a key to a more animal-like type of intelligence. Here, we propose a continual generalization of the chunking problem (an unsupervised problem), encompassing fixed and probabilistic chunks, discovery of temporal and causal structures and their continual variations. Additionally, we propose an algorithm called SyncMap that can learn and adapt to changes in the problem by creating a dynamic map which preserves the correlation between variables. Results of SyncMap suggest that the proposed algorithm learn near optimal solutions, despite the presence of many types of structures and their continual variation. When compared to Word2vec, PARSER and MRIL, SyncMap surpasses or ties with the best algorithm on 66% of the scenarios while being the second best in the remaining 34%. SyncMap's model-free simple dynamics and the absence of loss functions reveal that, perhaps surprisingly, much can be done with self-organization alone.

  • Improving robustness for vision transformer with a simple dynamic scanning augmentation augmentation 査読 国際誌

    Shashank Kotyan, Danilo Vasconcellos Vargas

    Neurocomputing   2024年1月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Symmetrical SyncMap for imbalanced general chunking problems 査読 国際誌

    Heng Zhang , Danilo Vasconcellos Vargas

    Physica D: Nonlinear Phenomena   2023年12月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • “Magnum: Tackling High-Dimensional Structures with Self-Organization 査読 国際誌

    Po Yuan Mao, Yik Foong Tham, Heng Zhang, Danilo Vasconcellos Vargas

    Neurocomputing   2023年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Generating oscillation activity with Echo State Network to mimic the behaviour of a simple central pattern generator 査読 国際誌

    Yik Foong Tham, Danilo Vasconcellos Vargas

    Proceedings of the CogSci 2023.   2023年8月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • Curious-II: A Multi/Many-Objective Optimization Algorithm with Subpopulations based on Multi-novelty Search 査読 国際誌

    Yuzi Jiang, Danilo Vasconcellos Vargas

    GECCO '23 Companion: Proceedings of the Companion Conference on Genetic and Evolutionary Computation   2023年7月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • A survey on reservoir computing and its interdisciplinary applications beyond traditional machine learning 査読 国際誌

    Heng Zhang; Danilo Vasconcellos Vargas

    IEEE Access   2023年7月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

    Reservoir computing (RC), first applied to temporal signal processing, is a recurrent neural network in which neurons are randomly connected. Once initialized, the connection strengths remain unchanged. Such a simple structure turns RC into a non-linear dynamical system that maps low-dimensional inputs into a high-dimensional space. The model's rich dynamics, linear separability, and memory capacity then enable a simple linear readout to generate adequate responses for various applications. RC spans areas far beyond machine learning, since it has been shown that the complex dynamics can be realized in various physical hardware implementations and biological devices. This yields greater flexibility and shorter computation time. Moreover, the neuronal responses triggered by the model's dynamics shed light on understanding brain mechanisms that also exploit similar dynamical processes. While the literature on RC is vast and fragmented, here we conduct a unified review of RC's recent developments from machine learning to physics, biology, and neuroscience. We first review the early RC models, and then survey the state-of-the-art models and their applications. We further introduce studies on modeling the brain's mechanisms by RC. Finally, we offer new perspectives on RC development, including reservoir design, coding frameworks unification, physical RC implementations, and interaction between RC, cognitive neuroscience and evolution.

  • Image Generation from Text and Segmentation 査読 国際誌

    Masato Osugi, and Danilo Vasconcellos Vargas

    Proceedings of the 2022 Tenth International Symposium on Computing and Networking Workshops (CANDARW). IEEE   2022年8月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • Understanding SyncMap’s Dynamics and Its Self-organization Properties: A Space-time Analysis 査読 国際誌

    Zhang, Heng, and Danilo Vasconcellos Vargas

    Proceedings of the 2022 5th Artificial Intelligence and Cloud Computing Conference   2022年8月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • Understanding SyncMap: Analyzing the components of Its Dynamical Equation 査読 国際誌

    Yik Foong Tham and Danilo Vasconcellos Vargas

    Proceedings of the 2022 5th Artificial Intelligence and Cloud Computing Conference   2022年8月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • Transferability Of Features For Neural Networks Links To Adversarial Attacks and Defences 査読 国際誌

    S. Kotyan, M. Matsuki, and D. V. Vargas

    PLOS One   2022年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Adversarial Robustness Assessment: Why both L0 and L∞ Attacks Are Necessary 査読 国際誌

    S. Kotyan and D. V. Vargas,

    PLOS One   2022年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Towards Evaluating the Representation Learned by Variational AutoEncoders 査読 国際誌

    T. Ueda and D. V. Vargas

    SICE 2021   2021年7月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • “Preliminary results on Chunking with Recurrent Neural Networks” 査読 国際誌

    Po-Yuan Mao and Danilo Vasconcellos Vargas

    SICE 2021   2021年7月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • “Deep neural network loses attention to adversarial images” 査読 国際誌

    S. Kotyan and D. V. Vargas,

    AISafety Workshop (AISafety 2021)   2021年5月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • “Towards Understanding The Space of Unrobust Features of Neural Networks” 査読 国際誌

    L. Bingli, T. Kanzaki and D. V. Vargas,

    IEEE CYBCONF 2021   2021年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • “Setting the space for deliberation in decision-making” 査読 国際誌

    D. V. Vargas and J. Lauwereyns

    Cognitive Neurodynamics   2021年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

    Decision-making models in the behavioral, cognitive, and neural sciences typically consist of forced-choice paradigms with two alternatives. While theoretically it is feasible to translate any decision situation to a sequence of binary choices, real-life decision-making is typically more complex and nonlinear, involving choices among multiple items, graded judgments, and deferments of decision-making. Here, we discuss how the complexity of real-life decision-making can be addressed using conventional decision-making models by focusing on the interactive dynamics between criteria settings and the collection of evidence. Decision-makers can engage in multi-stage, parallel decision-making by exploiting the space for deliberation, with non-binary readings of evidence available at any point in time. The interactive dynamics principally adhere to the speed-accuracy tradeoff, such that increasing the space for deliberation enables extended data collection. The setting of space for deliberation reflects a form of meta-decision-making that can, and should be, studied empirically as a value-based exercise that weighs the prior propensities, the economics of information seeking, and the potential outcomes. Importantly, the control of the space for deliberation raises a question of agency. Decision-makers may actively and explicitly set their own decision parameters, but these parameters may also be set by environmental pressures. Thus, decision-makers may be influenced—or nudged in a particular direction—by how decision problems are framed, with a sense of urgency or a binary definition of choice options. We argue that a proper understanding of these mechanisms has important practical implications toward the optimal usage of space for deliberation.

  • “Preliminary Results for Subpopulation Algorithm Based on Novelty (SAN) Compared with the State of the Art” 査読 国際誌

    Y. Jiang (South China University of Technology) and D. V. Vargas,

    IEEE CYBCONF 2021   2021年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • “Parameter Optimization via CMA-ES for Implementation in the Active Control of Magnetic Pillar Arrays” 査読 国際誌

    S. Gaysornkaew, D. V. Vargas and F. Tsumori,

    IEEE CYBCONF 2021   2021年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • “Towards learning Hierarchical Structures with SyncMap” 査読 国際誌

    Y. F. Tham and D. V. Vargas,

    IEEE CYBCONF 2021   2021年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • Preliminary study of applied Binary Neural Networks for Neural Cryptography 査読

    Valencia R., Sham, B., D. V. Vargas

    GECCO 2020   2020年7月

     詳細を見る

    記述言語:英語  

  • Towards improvement of SUNA in Multiplexers with preliminary results of simple Logic Gate neuron variation 査読

    Anh Duc Ta, D. V. Vargas

    GECCO 2020   2020年7月

     詳細を見る

    記述言語:英語  

  • Towards Evolving Robust Neural Architectures to Defend from Adversarial Attacks 査読

    Shashank Kotayan, D. V. Vargas

    GECCO 2020   2020年7月

     詳細を見る

    記述言語:英語  

  • Is Neural Architecture Search A Way Forward to Develop Robust Neural Networks? 査読

    Shashank Kotayan, D. V. Vargas

    人工知能学会全国大会 (JSAI 2020)   2020年6月

     詳細を見る

    記述言語:英語  

  • Evolving Robust Neural Architectures to Defend from Adversarial Attacks 査読

    Shashank Kotayan, D. V. Vargas

    AIsafety workshop (AIsafety 2020)   2020年6月

     詳細を見る

    記述言語:英語  

  • Understanding the one-pixel attack: Propagation maps and locality analysis 査読

    D. V. Vargas, Jiawei Su

    AIsafety workshop (AIsafety 2020)   2020年6月

     詳細を見る

    記述言語:英語  

  • Adversarial Machine Learning A Blow to the Transportation Sharing Economy

    Steven Van Uytsel, Danilo Vasconcellos Vargas

    Perspectives in Law, Business and Innovation   179 - 208   2020年1月

     詳細を見る

    記述言語:英語  

    Adversarial machine learning has indicated that perturbations to a picture may disable a deep neural network from correctly qualifying the content of a picture. The progressing research has even revealed that the perturbations do not necessarily have to be large in size. This research has been transplanted to traffic signs. The test results were disastrous. For example, a perturbated stop sign was recognized as a speeding sign. Because visualization technology is not able to overcome this problem yet, the question arises who should be liable for accidents caused by this technology. Manufacturers are being pointed at and for that reason it has been claimed that the commercialization of autonomous vehicles may stall. Without autonomous vehicles, the sharing economy may not fully develop either. This chapter shows that there are alternatives for the unpredictable financial burden on the car manufacturers for accidents with autonomous cars. This chapter refers to operator liability, but argues that for reasons of fairness, this is not a viable choice. A more viable choice is a no-fault liability on the manufacturer, as this kind of scheme forces the car manufacturer to be careful but keeps the financial risk predicable. Another option is to be found outside law. Engineers could build infrastructure enabling automation. Such infrastructure may overcome the problems of the visualization technology, but could potentially create a complex web of product and service providers. Legislators should prevent that the victims of an accident, if it were still to occur, would face years in court with the various actors of this complex web in order to receive compensation.

    DOI: 10.1007/978-981-15-1350-3_11

  • One-Pixel Attack: Understanding and Improving Deep Neural Networks with Evolutionary Computation. In Deep Neural Evolution 国際誌

    Danilo Vasconcellos Vargas

    Deep Neural Evolution   2020年1月

     詳細を見る

    記述言語:英語  

    その他リンク: https://www.springer.com/gp/book/9789811536847

  • One Pixel Attack for Fooling Deep Neural Networks 査読

    Jiawei Su, Danilo Vasconcellos Vargas, Kouichi Sakurai

    IEEE Transactions on Evolutionary Computation   23 ( 5 )   828 - 841   2019年10月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

    Recent research has revealed that the output of deep neural networks (DNNs) can be easily altered by adding relatively small perturbations to the input vector. In this paper, we analyze an attack in an extremely limited scenario where only one pixel can be modified. For that we propose a novel method for generating one-pixel adversarial perturbations based on differential evolution (DE). It requires less adversarial information (a black-box attack) and can fool more types of networks due to the inherent features of DE. The results show that 67.97% of the natural images in Kaggle CIFAR-10 test dataset and 16.04% of the ImageNet (ILSVRC 2012) test images can be perturbed to at least one target class by modifying just one pixel with 74.03% and 22.91% confidence on average. We also show the same vulnerability on the original CIFAR-10 dataset. Thus, the proposed attack explores a different take on adversarial machine learning in an extreme limited scenario, showing that current DNNs are also vulnerable to such low dimension attacks. Besides, we also illustrate an important application of DE (or broadly speaking, evolutionary computation) in the domain of adversarial machine learning: creating tools that can effectively generate low-cost adversarial attacks against neural networks for evaluating robustness.

    DOI: 10.1109/TEVC.2019.2890858

  • Linear Subspace Paradigm: An Investigation on a Novel Paradigm for Optimizing Deep Neural Networks 査読 国際誌

    Lia T. Parsenadze, Danilo Vasconcellos Vargas, Toshiyuki Fujita

    EUROGEN 2019   2019年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • Universal Rules for Fooling Deep Neural Networks based Text Classification 招待 査読 国際誌

    Di Li, Danilo Vasconcellos Vargas, Sakurai Kouichi

    In IEEE Congress on Evolutionary Computation (CEC) 2019   2019年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • Batch Tournament Selection for Genetic Programming 招待 査読 国際誌

    Vinicius Veloso de Melo, Danilo Vasconcellos Vargas, Wolfgang Banzhaf

    GECCO 2019   2019年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Towards Evolutionary-based Classifiers Implemented with an Optical Fluorescent Voxels System 招待 査読 国際誌

    Danilo Vasconcellos Vargas, Hiroaki Yoshioka, Daisuke Nakamura, Takatsugu Ono, Naoya Tate

    Optics and Photonics International Congress 2019   2019年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Towards Solving Neural Networks with Optimization Trajectory Search 招待 査読 国際誌

    Lia T. Parsenadze, Danilo Vasconcellos Vargas, Toshiyuki Fujita

    GECCO (Companion) 2019   2019年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Self-Training Autonomous Driving Agent 査読 国際誌

    Shashank Kotayan, Danilo Vasconcellos Vargas, Venkanna U

    SICE 2019   2019年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(国際会議プロシーディングス)  

  • Attacking convolutional neural network using differential evolution 査読

    Jiawei Su, Danilo Vasconcellos Vargas, Kouichi Sakurai

    IPSJ Transactions on Computer Vision and Applications   11 ( 1 )   2019年2月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

    The output of convolutional neural networks (CNNs) has been shown to be discontinuous which can make the CNN image classifier vulnerable to small well-tuned artificial perturbation. That is, images modified by conducting such alteration (i.e., adversarial perturbation) that make little difference to the human eyes can completely change the CNN classification results. In this paper, we propose a practical attack using differential evolution (DE) for generating effective adversarial perturbations. We comprehensively evaluate the effectiveness of different types of DEs for conducting the attack on different network structures. The proposed method only modifies five pixels (i.e., few-pixel attack), and it is a black-box attack which only requires the miracle feedback of the target CNN systems. The results show that under strict constraints which simultaneously control the number of pixels changed and overall perturbation strength, attacking can achieve 72.29%, 72.30%, and 61.28% non-targeted attack success rates, with 88.68%, 83.63%, and 73.07% confidence on average, on three common types of CNNs. The attack only requires modifying five pixels with 20.44, 14.28, and 22.98 pixel value distortion. Thus, we show that current deep neural networks are also vulnerable to such simpler black-box attacks even under very limited attack conditions.

    DOI: 10.1186/s41074-019-0053-3

  • A new design for evaluating moving target defense system 査読

    Wai Kyi Kyi Oo, Hiroshi Koide, Danilo Vasconcellos Vargas, Kouichi Sakurai

    6th International Symposium on Computing and Networking Workshops, CANDARW 2018 Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018   561 - 563   2018年12月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Moving Target Defense (MTD) concept has been a feasible idea for reducing the possibility of attack happening through alternation attack surfaces or diversification the attribute or parameters of a protected system. As a result of applying MTD techniques to the system, an attacker would have more difficulties in exploiting a vulnerabilities of the target system. This study proposes an evaluation method of MTD systems combined with several different MTD techniques. The proposed method is a primary step in designing an evaluation model for the effectiveness of MTD. The main goal is to estimate the attack success ratio on the MTD systems mitigating from threats of executable binary file or malware injection. With the proposed evaluation method, we expect to prove that the MTD technology can enhance the security of a web server, and can be applied in a real-world information system. As our preliminary work done, we set up a prototype framework to validate the proposed work in a pseudo-experimental environment.

    DOI: 10.1109/CANDARW.2018.00111

  • Tracing MIRAI malware in networked system 査読

    Yao Xu, Hiroshi Koide, Danilo Vasconcellos Vargas, Kouichi Sakurai

    6th International Symposium on Computing and Networking Workshops, CANDARW 2018 Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018   534 - 538   2018年12月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    In 2021, it is anticipated that there will be approximately 30 billion Internet of Things (IoT) devices. The tremendous aggregate value of the IoT makes it a tempting and lucrative target for cyber criminals. The breakout of Mirai malware, which compromises poorly secured IoT devices with factory-default username and passphrase to launch Distributed Denial of Service (DDoS) attacks, has raised broad awareness towards the need for increased IoT security. To better defend against Mirai infection and spread, it is critical to know how the malware operates as the first step. In this paper, we give a combined static and dynamic analysis of Mirai, basing on the results of which, we introduce the application of Threat Tracer. Threat tracer is an information system simulator initially developed to help design a system robust against Advanced Persistent Attacks(APT). It offers an intuitive track on how a cyber threat behaves in a complicated networked system. The feedback simultaneously contributes to revealing vulnerabilities of a system. Our work focuses on the replication of Mirai Malware's operating processes in Threat Tracer simulation. By achieving doing so, we believe it could offer a comprehensible description of how Mirai acts. Also, considering the continuous emergence of Mirai variants, the simulation serves as a predictor on upcoming threats' behavior patterns.

    DOI: 10.1109/CANDARW.2018.00104

  • Neural cryptography based on the topology evolving neural networks 査読

    Yuetong Zhu, Danilo Vasconcellos Vargas, Kouichi Sakurai

    6th International Symposium on Computing and Networking Workshops, CANDARW 2018 Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018   472 - 478   2018年12月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Modern cryptographic schemes is developed based on the mathematical theory. Recently works show a new direction about cryptography based on the neural networks. Instead of learning a specific algorithm, a cryptographic scheme is generated automatically. While one kind of neural network is used to achieve the scheme, the idea of the neural cryptography can be realized by other neural network architecture is unknown. In this paper, we make use of this property to create neural cryptography scheme on a new topology evolving neural network architecture called Spectrum-diverse unified neuroevolution architecture. First, experiments are conducted to verify that Spectrum-diverse unified neuroevolution architecture is able to achieve automatic encryption and decryption. Subsequently, we do experiments to achieve the neural symmetric cryptosystem by using adversarial training.

    DOI: 10.1109/CANDARW.2018.00091

  • Empirical evaluation on robustness of deep convolutional neural networks activation functions against adversarial perturbation 査読

    Jiawei Su, Danilo Vasconcellos Vargas, Kouichi Sakurai

    6th International Symposium on Computing and Networking Workshops, CANDARW 2018 Proceedings - 2018 6th International Symposium on Computing and Networking Workshops, CANDARW 2018   223 - 227   2018年12月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Recent research has shown that deep convolutional neural networks (DCNN) are vulnerable to several different types of attacks while the reasons of such vulnerability are still under investigation. For instance, the adversarial perturbations can conduct a slight change on a natural image to make the target DCNN make the wrong recognition, while the reasons that DCNN is sensitive to such small modification are divergent from one research to another. In this paper, we evaluate the robustness of two commonly used activation functions of DCNN, namely the sigmoid and ReLu, against the recently proposed low-dimensional one-pixel attack. We show that the choosing of activation functions can be an important factor that influences the robustness of DCNN. The results show that comparing with sigmoid, the ReLu non-linearity is more vulnerable which allows the low dimensional one-pixel attack exploit much higher success rate and confidence of launching the attack. The results give insights on designing new activation functions to enhance the security of DCNN.

    DOI: 10.1109/CANDARW.2018.00049

  • Introducing learning classifier systems Rules that capture complexity

    Ryan J. Urbanowicz, Danilo Vasconcellos Vargas

    2018 Genetic and Evolutionary Computation Conference, GECCO 2018 GECCO 2018 Companion - Proceedings of the 2018 Genetic and Evolutionary Computation Conference Companion   619 - 648   2018年7月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    DOI: 10.1145/3205651.3207869

  • Lightweight Classification of IoT Malware Based on Image Recognition 査読

    Jiawei Su, Danilo Vasconcellos Vargas, Sanjiva Prasad, Sgandurra Daniele, Yaokai Feng, Kouichi Sakurai

    42nd IEEE Computer Software and Applications Conference, COMPSAC 2018 Proceedings - 2018 IEEE 42nd Annual Computer Software and Applications Conference, COMPSAC 2018   664 - 669   2018年6月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    The Internet of Things (IoT) is an extension of the traditional Internet, which allows a very large number of smart devices, such as home appliances, network cameras, sensors and controllers to connect to one another to share information and improve user experiences. IoT devices are micro-computers for domain-specific computations rather than traditional functionspecific embedded devices. This opens the possibility of seeing many kinds of existing attacks, traditionally targeted at the Internet, also directed at IoT devices. As shown by recent events, such as the Mirai and Brickerbot botnets, DDoS attacks have become very common in IoT environments as these lack basic security monitoring and protection mechanisms. In this paper, we propose a novel light-weight approach for detecting DDos malware in IoT environments. We extract the malware images (i.e., a one-channel gray-scale image converted from a malware binary) and utilize a light-weight convolutional neural network for classifying their families. The experimental results show that the proposed system can achieve 94:0% accuracy for the classification of goodware and DDoS malware, and 81:8% accuracy for the classification of goodware and two main malware families.

    DOI: 10.1109/COMPSAC.2018.10315

  • Effectively Protect Your Privacy Enabling Flexible Privacy Control on Web Tracking 査読

    Shiqian Yu, Danilo Vasconcellos Vargas, Kouichi Sakurai

    5th International Symposium on Computing and Networking, CANDAR 2017 Proceedings - 2017 5th International Symposium on Computing and Networking, CANDAR 2017   533 - 536   2018年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Third-party tracking, which can collect the users' privacy when users are surfing the Internet, has garnered much attention. Nowadays tracker-blocking tools often use a ruleset based on the domains and elements that need to be blocked. This results in blocking all access tracking, even though the website shows no sign about tracking users' privacy. And what's more, although the tracker-blocking tools try their best to block all the third-party tracking, not all the users dislike the advertisement. Some of them think if their privacy is fine, it's all right to accept advertisements. In this paper, we present a novel framework by using Word2Vec to block third-party tracking. Our goal is to create more flexible and well-developed ruleset that can help users to protect their privacy according to their needs. Instead of blocking all access tracking, we decide to pay more attention to the websites that have a strong probability to collect the users' privacy. We use Word2Vec to classify the websites, and our results show that after using our framework, the error rate drops from 71% to 24%. We believe it brings the new blood into the field of web privacy by providing not only the new third-party tracking tool but also a novel way of thinking about how to block the third-party tracking.

    DOI: 10.1109/CANDAR.2017.26

  • Spectrum-diverse neuroevolution with unified neural models 査読

    Danilo Vasconcellos Vargas, Junichi Murata

    IEEE Transactions on Neural Networks and Learning Systems   28 ( 8 )   1759 - 1773   2017年8月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

    Learning algorithms are being increasingly adopted in various applications. However, further expansion will require methods that work more automatically. To enable this level of automation, a more powerful solution representation is needed. However, by increasing the representation complexity, a second problem arises. The search space becomes huge, and therefore, an associated scalable and efficient searching algorithm is also required. To solve both the problems, first a powerful representation is proposed that unifies most of the neural networks features from the literature into one representation. Second, a new diversity preserving method called spectrum diversity is created based on the new concept of chromosome spectrum that creates a spectrum out of the characteristics and frequency of alleles in a chromosome. The combination of spectrum diversity with a unified neuron representation enables the algorithm to either surpass or equal NeuroEvolution of Augmenting Topologies on all of the five classes of problems tested. Ablation tests justify the good results, showing the importance of added new features in the unified neuron representation. Part of the success is attributed to the novelty-focused evolution and good scalability with a chromosome size provided by spectrum diversity. Thus, this paper sheds light on a new representation and diversity preserving mechanism that should impact algorithms and applications to come.

    DOI: 10.1109/TNNLS.2016.2551748

  • A Brief Review on Anomaly Detection and its Applications to Cybersecurity (情報通信システムセキュリティ)

    Danilo Vasconcellos Vargas, Kouichi Sakurai

    IEICE technical report   116 ( 522 )   37 - 42   2017年3月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Evasion attacks against statistical code obfuscation detectors 査読

    Jiawei Su, Danilo Vasconcellos Vargas, Kouichi Sakurai

    12th International Workshop on Security, IWSEC 2017 Advances in Information and Computer Security - 12th International Workshop on Security, IWSEC 2017, Proceedings   121 - 137   2017年1月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    In the domain of information security, code obfuscation is a feature often employed for malicious purposes. For example there have been quite a few papers reporting that obfuscated JavaScript frequently comes with malicious functionality such as redirecting to external malicious websites. In order to capture such obfuscation, a class of detectors based on statistical features of code, mostly n-grams have been proposed and been claimed to achieve high detection accuracy. In this paper, we formalize a common scenario between defenders who maintain the statistical obfuscation detectors and adversaries who want to evade the detection. Accordingly, we create two kinds of evasion attack methods and evaluate the robustness of statistical detectors under such attacks. Experimental results show that statistical obfuscation detectors can be easily fooled by a sophisticated adversary even in worst case scenarios.

    DOI: 10.1007/978-3-319-64200-0_8

  • Curious Searching for unknown regions of space with a subpopulation-based algorithm 査読

    Danilo Vasconcellos Vargas, Junichi Murata

    2016 Genetic and Evolutionary Computation Conference, GECCO 2016 Companion GECCO 2016 Companion - Proceedings of the 2016 Genetic and Evolutionary Computation Conference   145 - 146   2016年7月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Intrinsic motivation and novelty search are promising approaches to deal with plateaus, deceptive functions and other exploration problems where using only the main objective function is insufficient. However, it is not clear until now how and if intrinsic motivation (novelty search) can improve single objective algorithms in general. The hurdle is that using multi-objective algorithms to deal with single-objective problems adds an unnecessary overhead such as the search for non-dominated solutions. Here, we propose the Curious algorithm which is the first multi-objective algorithm focused on solving single-objective problems. Curious uses two subpopulations algorithms. One subpopulation is dedicated for improving objective function values and another one is added to search for unknown regions of space based on objective prediction errors. By using a differential evolution operator, genes from individuals in all subpopulations are mixed. In this way, the promising regions (solutions with high fitness) and unknown regions (solutions with high prediction error) are searched simultaneously. Because of thus realized strong yet well controlled novelty search, the algorithm possesses powerful exploration ability and outperforms usual single population based algorithms such as differential evolution. Thus, it demonstrates that the addition of intrinsic motivation is promising and should improve further single objective algorithms in general.

    DOI: 10.1145/2908961.2908982

  • Spectrum-diverse neuroevolution with unified neural models 査読

    Danilo Vasconcellos Vargas, Junichi Murata

    IEEE Transactions on Neural Networks and Learning Systems   28 ( 8 )   1759 - 1773   2016年

     詳細を見る

    記述言語:その他   掲載種別:研究論文(学術雑誌)  

  • Novelty-Organizing Team of Classifiers in noisy and dynamic environments 査読

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    IEEE Congress on Evolutionary Computation, CEC 2015 2015 IEEE Congress on Evolutionary Computation, CEC 2015 - Proceedings   2937 - 2944   2015年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    In the real world, the environment is constantly changing with the input variables under the effect of noise. However, few algorithms were shown to be able to work under those circumstances. Here, Novelty-Organizing Team of Classifiers (NOTC) is applied to the continuous action mountain car as well as two variations of it: A noisy mountain car and an unstable weather mountain car. These problems take respectively noise and change of problem dynamics into account. Moreover, NOTC is compared with NeuroEvolution of Augmenting Topologies (NEAT) in these problems, revealing a trade-off between the approaches. While NOTC achieves the best performance in all of the problems, NEAT needs less trials to converge. It is demonstrated that NOTC achieves better performance because of its division of the input space (creating easier problems). Unfortunately, this division of input space also requires a bit of time to bootstrap.

    DOI: 10.1109/CEC.2015.7257254

  • The relationship between (Un)fractured problems and division of input space

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    17th Genetic and Evolutionary Computation Conference, GECCO 2015 GECCO 2015 - Companion Publication of the 2015 Genetic and Evolutionary Computation Conference   981 - 987   2015年7月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Problems can be categorized as fractured or unfractured ones. A different set of characteristics are needed for learning algorithms to solve each of these two types of problems. However, the exact characteristics needed to solve each type are unclear. This article shows that the division of the input space is one of these characteristics. In other words, a study is presented showing that while fractured problems benefit from a finer division of the input space, unfractured problems benefit from a coarser division of input space. Many open questions still remains. And the article discusses two conjectures which can be used to solve fractured problems more easily.

    DOI: 10.1145/2739482.2768447

  • General subpopulation framework and taming the conflict inside populations 査読

    Danilo Vasconcellos Vargas, Junichi Murata, Hirotaka Takano, Alexandre Cláudio Botazzo Delbem

    Evolutionary Computation   23 ( 1 )   1 - 36   2015年3月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

    Structured evolutionary algorithms have been investigated for some time. However, they have been under explored especially in the field of multi-objective optimization. Despite good results, the use of complex dynamics and structures keep the understanding and adoption rate of structured evolutionary algorithms low. Here, we propose a general subpopulation framework that has the capability of integrating optimization algorithms without restrictions as well as aiding the design of structured algorithms. The proposed framework is capable of generalizing most of the structured evolutionary algorithms, such as cellular algorithms, island models, spatial predator-prey, and restricted mating based algorithms. Moreover, we propose two algorithms based on the general subpopulation framework, demonstrating that with the simple addition of a number of single-objective differential evolution algorithms for each objective, the results improve greatly, even when the combined algorithms behave poorly when evaluated alone at the tests. Most importantly, the comparison between the subpopulation algorithms and their related panmictic algorithms suggests that the competition between different strategies inside one population can have deleterious consequences for an algorithm and reveals a strong benefit of using the subpopulation framework.

    DOI: 10.1162/EVCO_a_00118

  • Novelty-organizing team of classifiers - A team-individual multi-objective approach to reinforcement learning

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2014 53rd Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2014 Proceedings of the SICE Annual Conference   1785 - 1792   2014年10月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    In reinforcement learning, there are basically two spaces to search: value-function space and policy space. Consequently, there are two fitness functions each with their associated trade-offs. However, the problem is still perceived as a single-objective one. Here a multi-objective reinforcement learning algorithm is proposed with a structured novelty map population evolving feedforward neural models. It outperforms a gradient based continuous input-output state-of-art algorithm in two problems. Contrary to the gradient based algorithm, the proposed one solves both problems with the same parameters and smaller variance of results. Moreover, the results are comparable even with other discrete action algorithms of the literature as well as neuroevolution methods such as NEAT. The proposed method brings also the novelty map population concept, i.e., a novelty map-based population which is less sensitive to the input distribution and therefore more suitable to create the state space. In fact, the novelty map framework is shown to be less dynamic and more resource efficient than variants of the self-organizing map.

    DOI: 10.1109/SICE.2014.6935299

  • Novelty-Organizing Classifiers applied to classification and reinforcement learning Towards flexible algorithms

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    16th Genetic and Evolutionary Computation Conference, GECCO 2014 GECCO 2014 - Companion Publication of the 2014 Genetic and Evolutionary Computation Conference   81 - 82   2014年

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    It is widely known that reinforcement learning is a more general problem than supervised learning. In fact, supervised learning can be seen as a class of reinforcement learning problems. However, only a couple of papers tested reinforcement learning algorithms in supervised learning problems. Here we propose a new and simpler way to abstract supervised learning for any reinforcement learning algorithm. Moreover, a new algorithm called Novelty-Organizing Classifiers is developed based on a Novelty Map population that focuses more on the novelty of the inputs than their frequency. A comparison of the proposed method with Self-Organizing Classifiers and BioHel on some datasets is presented. Even though BioHel is specialized in solving supervised learning problems, the results showed only a trade-off between the algorithms. Lastly, results on a maze problem validate the flexibility of the proposed algorithm beyond supervised learning problems. Thus, Novelty-Organizing Classifiers is capable of solving many supervised learning problems as well as a maze problem without changing any parameter at all. Considering the fact that no adaptation of parameters was executed, the proposed algorithm's basis seems interestingly flexible.

    DOI: 10.1145/2598394.2598429

  • Phylogenetic differential evolution

    Vinicius Veloso de Melo, Danilo Vasconcellos Vargas, Marcio Kassouf Crocomo

    Natural Computing for Simulation and Knowledge Discovery   22 - 40   2014年

     詳細を見る

    記述言語:その他  

  • Self organizing classifiers First steps in structured evolutionary machine learning 査読

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    Evolutionary Intelligence   6 ( 2 )   57 - 72   2013年11月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

    Learning classifier systems (LCSs) are evolutionary machine learning algorithms, flexible enough to be applied to reinforcement, supervised and unsupervised learning problems with good performance. Recently, self organizing classifiers were proposed which are similar to LCSs but have the advantage that in its structured population no balance between niching and fitness pressure is necessary. However, more tests and analysis are required to verify its benefits. Here, a variation of the first algorithm is proposed which uses a parameterless self organizing map (SOM). This algorithm is applied in challenging problems such as big, noisy as well as dynamically changing continuous input-action mazes (growing and compressing mazes are included) with good performance. Moreover, a genetic operator is proposed which utilizes the topological information of the SOM's population structure, improving the results. Thus, the first steps in structured evolutionary machine learning are shown, nonetheless, the problems faced are more difficult than the state-of-art continuous input-action multi-step ones.

    DOI: 10.1007/s12065-013-0095-x

  • Self organizing classifiers and niched fitness

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2013 15th Genetic and Evolutionary Computation Conference, GECCO 2013 GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference   1109 - 1116   2013年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Learning classifier systems are adaptive learning systems which have been widely applied in a multitude of application domains. However, there are still some generalization problems unsolved. The hurdle is that fitness and niching pressures are difficult to balance. Here, a new algorithm called Self Organizing Classifiers is proposed which faces this problem from a different perspective. Instead of balancing the pressures, both pressures are separated and no balance is necessary. In fact, the proposed algorithm possesses a dynamical population structure that self-organizes itself to better project the input space into a map. The niched fitness concept is defined along with its dynamical population structure, both are indispensable for the understanding of the proposed method. Promising results are shown on two continuous multi-step problems. One of which is yet more challenging than previous problems of this class in the literature.

    DOI: 10.1145/2463372.2463501

  • The Motivation behind Low Dimensional Weak Learners Extends to Ensembles of Decision Trees (システム研究会 インテリジェント・システム・シンポジウム(FANシンポジウム))

    Danilo Vasconcellos Vargas, 高野 浩貴, 村田 純一

    電気学会研究会資料. SA, 静止器研究会   2013 ( 29 )   23 - 26   2013年9月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(学術雑誌)  

  • Phylogenetic Differential Evolution

    Vinícius Veloso de Melo, Danilo Vasconcellos Vargas, Marcio Kassouf Crocomo

    Natural Computing for Simulation and Knowledge Discovery   22 - 40   2013年7月

     詳細を見る

    記述言語:英語  

    This paper presents a new technique for optimizing binary problems with building blocks. The authors have developed a different approach to existing Estimation of Distribution Algorithms (EDAs). Our technique, called Phylogenetic Differential Evolution (PhyDE), combines the Phylogenetic Algorithm and the Differential Evolution Algorithm. The first one is employed to identify the building blocks and to generate metavariables. The second one is used to find the best instance of each metavariable. In contrast to existing EDAs that identify the related variables at each iteration, the presented technique finds the related variables only once at the beginning of the algorithm, and not through the generations. This paper shows that the proposed technique is more efficient than the well known EDA called Extended Compact Genetic Algorithm (ECGA), especially for large-scale systems which are commonly found in real world problems.

    DOI: 10.4018/978-1-4666-4253-9.ch002

  • A study on the importance of selection pressure and low dimensional weak learners to produce robust ensembles

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    15th Annual Conference on Genetic and Evolutionary Computation, GECCO 2013 GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference Companion   1755 - 1756   2013年

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Ensembles of classifiers have been studied for some time. It is widely known that weak learners should be accurate and diverse. However, in the real world there are many constraints and few have been said about the robustness of ensembles and how to develop it. In the context of ran- dom subspace methods, this paper addresses the question of developing ensembles to face problems under time con- straints. Experiments show that selecting weak learners based on their accuracy can be used to create robust en- sembles. Thus, the selection pressure in ensembles is a key technique to create not just effective ensembles but also robust ones. Moreover, the experiments motivate further research on ensembles made of low dimensional classifiers which achieve general accurate results.

    DOI: 10.1145/2464576.2480775

  • Self organizing classifiers and niched fitness

    Danilo V Vargas, Hirotaka Takano, Junichi Murata

    Proceedings of the 15th annual conference on Genetic and evolutionary computation   1109 - 1116   2013年

     詳細を見る

    記述言語:その他   掲載種別:研究論文(その他学術会議資料等)  

  • Continuous adaptive reinforcement learning with the evolution of Self Organizing Classifiers

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013 - Electronic Conference Proceedings   2013年

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    Learning classifier systems have been solving reinforcement learning problems for some time. However, they face difficulties under multi-step continuous problems. Adaptation may also become harder with time since the convergence of the population decreases its diversity. This article demonstrate that the novel Self Organizing Classifiers method can cope with dynamical multi-step continuous problems. Moreover, adaptation remains the same after convergence.

    DOI: 10.1109/DevLrn.2013.6652558

  • Contingency training

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2013 52nd Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2013 SICE 2013: International Conference on Instrumentation, Control, Information Technology and System Integration - SICE Annual Conference 2013, Conference Proceedings   1361 - 1366   2013年

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    When applied to high-dimensional datasets, feature selection algorithms might still leave dozens of irrelevant variables in the dataset. Therefore, even after feature selection has been applied, classifiers must be prepared to the presence of irrelevant variables. This paper investigates a new training method called Contingency Training which increases the accuracy as well as the robustness against irrelevant attributes. Contingency training is classifier independent. By subsampling and removing information from each sample, it creates a set of constraints. These constraints aid the method to automatically find proper importance weights of the dataset's features. Experiments are conducted with the contingency training applied to neural networks over traditional datasets as well as datasets with additional irrelevant variables. For all of the tests, contingency training surpassed the unmodified training on datasets with irrelevant variables and even outperformed slightly when only a few or no irrelevant variables were present.

  • Multi-objective phylogenetic algorithm Solving multi-objective decomposable deceptive problems

    Jean Paulo Martins, Antonio Helson Mineiro Soares, Danilo Vasconcellos Vargas, Alexandre Cláudio Botazzo Delbem

    6th International Conference on Evolutionary Multi-Criterion Optimization, EMO 2011 Evolutionary Multi-Criterion Optimization - 6th International Conference, EMO 2011, Proceedings   285 - 297   2011年4月

     詳細を見る

    記述言語:英語   掲載種別:研究論文(その他学術会議資料等)  

    In general, Multi-objective Evolutionary Algorithms do not guarantee find solutions in the Pareto-optimal set. We propose a new approach for solving decomposable deceptive multi-objective problems that can find all solutions of the Pareto-optimal set. Basically, the proposed approach starts by decomposing the problem into subproblems and, then, combining the found solutions. The resultant approach is a Multi-objective Estimation of Distribution Algorithm for solving relatively complex multi-objective decomposable problems, using a probabilistic model based on a phylogenetic tree. The results show that, for the tested problem, the algorithm can efficiently find all the solutions of the Pareto-optimal set, with better scaling than the hierarchical Bayesian Optimization Algorithm and other algorithms of the state of art.

    DOI: 10.1007/978-3-642-19893-9_20

  • Multi-objective phylogenetic algorithm: Solving multi-objective decomposable deceptive problems

    Jean Paulo Martins, Antonio Helson Mineiro Soares, Danilo Vasconcellos Vargas, Alexandre Cláudio Botazzo Delbem

    International Conference on Evolutionary Multi-Criterion Optimization   285 - 297   2011年

     詳細を見る

    記述言語:その他   掲載種別:研究論文(その他学術会議資料等)  

▼全件表示

書籍等出版物

  • Autonomous Vehicles: Business, Technology and Law (Perspectives in Law, Business and Innovation)

    Steven Van Uytsel, Danilo Vasconcellos Vargas(担当:編集)

    Springer  2021年1月 

     詳細を見る

    記述言語:英語   著書種別:学術書

講演・口頭発表等

  • Understanding the one-pixel attack: Propagation maps and locality analysis 国際会議

    Vargas, D. V., & Su, J

    IJCAI2020 Workshop AISafety 2020  2020年7月 

     詳細を見る

    開催年月日: 2020年6月

    記述言語:英語   会議種別:シンポジウム・ワークショップ パネル(公募)  

    国名:日本国  

  • Is Neural Architecture Search A Way Forward to Develop Robust Neural Networks?

    Kotyan, S., Vargas, D. V.

    JSAI2020  2020年7月 

     詳細を見る

    開催年月日: 2020年6月

    記述言語:英語   会議種別:口頭発表(一般)  

    国名:日本国  

  • Towards improvement of SUNA in Multiplexers with preliminary results of simple Logic Gate neuron variation. 国際会議

    Anh Duc Ta, Vargas, D. V.

    GECCO2020 Late Breaking Papers (Companion)  2020年7月 

     詳細を見る

    開催年月日: 2020年6月

    記述言語:英語   会議種別:シンポジウム・ワークショップ パネル(公募)  

    国名:日本国  

  • Towards Evolving Robust Neural Architectures to Defend from Adversarial Attacks. 国際会議

    Kotyan, S., Vargas, D. V.

    GECCO2020 (Companion)  2020年7月 

     詳細を見る

    開催年月日: 2020年6月

    記述言語:英語   会議種別:口頭発表(一般)  

    国名:日本国  

  • Preliminary study of applied Binary Neural Networks for Neural Cryptography. 国際会議

    Valencia R., Sham, B., Vargas, D. V.

    GECCO2020 (Companion)  2020年7月 

     詳細を見る

    開催年月日: 2020年6月

    記述言語:英語   会議種別:口頭発表(一般)  

    国名:日本国  

  • Towards Evolving Robust Neural Architectures to Defend from Adversarial Attacks 国際会議

    Kotyan, S., Vargas, D. V.

    IJCAI2020 Workshop AISafety 2020  2020年7月 

     詳細を見る

    開催年月日: 2020年6月

    記述言語:英語   会議種別:シンポジウム・ワークショップ パネル(公募)  

    国名:日本国  

  • Frame difference generative adversarial networks Clearer contour video generating

    Rui Qiu, Danilo Vasconcellos Vargas, Kouich Sakurai

    7th International Symposium on Computing and Networking Workshops, CANDARW 2019  2019年11月 

     詳細を見る

    開催年月日: 2019年11月

    記述言語:英語  

    開催地:Nagasaki   国名:日本国  

    Generating image and video is a hot topic in Deep Learning. Especially, generating video is a difficult but meaningful work. How to generate video which has diversity and plausibility is still a problem to be solved. In this paper, we propose a novel model of Generative Adversarial Network(GAN) which called FDGAN to generate clear contour lines. Unlike existing GAN that only use frames, our method extends to use inter-frame difference. First introduce two temporal difference methods to process the inter-frame. Then increase a frame difference discriminator to discriminate whether the inter-frame is true or not. Using the model and new structure proposed, we perform video generation experiments on several widely used benchmark datasets such as MOVING MNIST, UCF-101. Consequently, the results achieve state-of-the-art performance for clarifying contour lines. Both quantitative and qualitative evaluations were made to show the effectiveness of our methods.

  • Self Training Autonomous Driving Agent

    Shashank Kotyan, Danilo Vasconcellos Vargas, U. Venkanna

    58th Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2019  2019年9月 

     詳細を見る

    開催年月日: 2019年9月

    記述言語:英語  

    開催地:Hiroshima   国名:日本国  

    Intrinsically, driving is a Markov Decision Process which suits well the reinforcement learning paradigm. In this paper, we propose a novel agent which learns to drive a vehicle without any human assistance. We use the concept of reinforcement learning and evolutionary strategies to train our agent in a 2D simulation environment. Our model's architecture goes beyond the World Model's by introducing difference images in the autoencoder. This novel involvement of difference images in the auto-encoder gives a better representation of the latent space concerning the motion of the vehicle and helps an autonomous agent to learn more efficiently how to drive a vehicle. Results show that our method requires fewer (96% less) total agents, (87.5% less) agents per generations, (70% less) generations and (90% less) rollouts than the original architecture while achieving the same accuracy of the original.

  • Batch tournament selection for genetic programming

    Vinícius V. De Melo, Danilo Vasconcellos Vargas, Wolfgang Banzhaf

    2019 Genetic and Evolutionary Computation Conference, GECCO 2019  2019年7月 

     詳細を見る

    開催年月日: 2019年7月

    記述言語:英語  

    開催地:Prague   国名:チェコ共和国  

    Lexicase selection achieves very good solution quality by introducing ordered test cases. However, the computational complexity of lexicase selection can prohibit its use in many applications. In this paper, we introduce Batch Tournament Selection (BTS), a hybrid of tournament and lexicase selection which is approximately one order of magnitude faster than lexicase selection while achieving a competitive quality of solutions. Tests on a number of regression datasets show that BTS compares well with lexicase selection in terms of mean absolute error while having a speed-up of up to 25 times. Surprisingly, BTS and lexicase selection have almost no difference in both diversity and performance. This reveals that batches and ordered test cases are completely different mechanisms which share the same general principle fostering the specialization of individuals. This work introduces an efficient algorithm that sheds light onto the main principles behind the success of lexicase, potentially opening up a new range of possibilities for algorithms to come.

  • Towards solving neural networks with optimization trajectory search

    Lia T. Parsenadze, Danilo Vasconcellos Vargas, Toshiyuki Fujita

    2019 Genetic and Evolutionary Computation Conference, GECCO 2019  2019年7月 

     詳細を見る

    開催年月日: 2019年7月

    記述言語:英語  

    開催地:Prague   国名:チェコ共和国  

    Modern gradient based optimization methods for deep neural networks demonstrate outstanding results on image classification tasks. However, methods that do not rely on gradient feedback fail to tackle deep network optimization. In the field of evolutionary computation, applying evolutionary algorithms directly to network weights remains to be an unresolved challenge. In this paper we examine a new framework for the evolution of deep nets. Based on the empirical analysis, we propose the use of linear sub-spaces of problems to search for promising optimization trajectories in parameter space, opposed to weight evolution. We show that linear sub-spaces of loss functions are sufficiently well-behaved to allow trajectory evaluation. Furthermore, we introduce fitness measure to show that it is possible to correctly categorize trajectories according to their distance from the optimal path. As such, this work introduces an alternative approach to evolutionary optimization of deep networks.

  • Universal Rules for Fooling Deep Neural Networks based Text Classification

    Di Li, Danilo Vasconcellos Vargas, Sakurai Kouichi

    2019 IEEE Congress on Evolutionary Computation, CEC 2019  2019年6月 

     詳細を見る

    開催年月日: 2019年6月

    記述言語:英語  

    開催地:Wellington   国名:ニュージーランド  

    Recently, deep learning based natural language processing techniques are being extensively used to deal with spam mail, censorship evaluation in social networks, among others. However, there is only a couple of works evaluating the vulnerabilities of such deep neural networks. Here, we go beyond attacks to investigate, for the first time, universal rules, i.e., rules that are sample agnostic and therefore could turn any text sample in an adversarial one. In fact, the universal rules do not use any information from the method itself (no information from the method, gradient information or training dataset information is used), making them black-box universal attacks. In other words, the universal rules are sample and method agnostic. By proposing a coevolutionary optimization algorithm we show that it is possible to create universal rules that can automatically craft imperceptible adversarial samples (only less than five perturbations which are close to misspelling are inserted in the text sample). A comparison with a random search algorithm further justifies the strength of the method. Thus, universal rules for fooling networks are here shown to exist. Hopefully, the results from this work will impact the development of yet more sample and model agnostic attacks as well as their defenses.

  • A new design for evaluating moving target defense system

    Wai Kyi Kyi Oo, Hiroshi Koide, Danilo Vasconcellos Vargas, Kouichi Sakurai

    6th International Symposium on Computing and Networking Workshops, CANDARW 2018  2018年12月 

     詳細を見る

    開催年月日: 2018年11月

    記述言語:英語  

    開催地:Takayama   国名:日本国  

    Moving Target Defense (MTD) concept has been a feasible idea for reducing the possibility of attack happening through alternation attack surfaces or diversification the attribute or parameters of a protected system. As a result of applying MTD techniques to the system, an attacker would have more difficulties in exploiting a vulnerabilities of the target system. This study proposes an evaluation method of MTD systems combined with several different MTD techniques. The proposed method is a primary step in designing an evaluation model for the effectiveness of MTD. The main goal is to estimate the attack success ratio on the MTD systems mitigating from threats of executable binary file or malware injection. With the proposed evaluation method, we expect to prove that the MTD technology can enhance the security of a web server, and can be applied in a real-world information system. As our preliminary work done, we set up a prototype framework to validate the proposed work in a pseudo-experimental environment.

  • Tracing MIRAI malware in networked system

    Yao Xu, Hiroshi Koide, Danilo Vasconcellos Vargas, Kouichi Sakurai

    6th International Symposium on Computing and Networking Workshops, CANDARW 2018  2018年12月 

     詳細を見る

    開催年月日: 2018年11月

    記述言語:英語  

    開催地:Takayama   国名:日本国  

    In 2021, it is anticipated that there will be approximately 30 billion Internet of Things (IoT) devices. The tremendous aggregate value of the IoT makes it a tempting and lucrative target for cyber criminals. The breakout of Mirai malware, which compromises poorly secured IoT devices with factory-default username and passphrase to launch Distributed Denial of Service (DDoS) attacks, has raised broad awareness towards the need for increased IoT security. To better defend against Mirai infection and spread, it is critical to know how the malware operates as the first step. In this paper, we give a combined static and dynamic analysis of Mirai, basing on the results of which, we introduce the application of Threat Tracer. Threat tracer is an information system simulator initially developed to help design a system robust against Advanced Persistent Attacks(APT). It offers an intuitive track on how a cyber threat behaves in a complicated networked system. The feedback simultaneously contributes to revealing vulnerabilities of a system. Our work focuses on the replication of Mirai Malware's operating processes in Threat Tracer simulation. By achieving doing so, we believe it could offer a comprehensible description of how Mirai acts. Also, considering the continuous emergence of Mirai variants, the simulation serves as a predictor on upcoming threats' behavior patterns.

  • Neural cryptography based on the topology evolving neural networks

    Yuetong Zhu, Danilo Vasconcellos Vargas, Kouichi Sakurai

    6th International Symposium on Computing and Networking Workshops, CANDARW 2018  2018年12月 

     詳細を見る

    開催年月日: 2018年11月

    記述言語:英語  

    開催地:Takayama   国名:日本国  

    Modern cryptographic schemes is developed based on the mathematical theory. Recently works show a new direction about cryptography based on the neural networks. Instead of learning a specific algorithm, a cryptographic scheme is generated automatically. While one kind of neural network is used to achieve the scheme, the idea of the neural cryptography can be realized by other neural network architecture is unknown. In this paper, we make use of this property to create neural cryptography scheme on a new topology evolving neural network architecture called Spectrum-diverse unified neuroevolution architecture. First, experiments are conducted to verify that Spectrum-diverse unified neuroevolution architecture is able to achieve automatic encryption and decryption. Subsequently, we do experiments to achieve the neural symmetric cryptosystem by using adversarial training.

  • Empirical evaluation on robustness of deep convolutional neural networks activation functions against adversarial perturbation

    Jiawei Su, Danilo Vasconcellos Vargas, Kouichi Sakurai

    6th International Symposium on Computing and Networking Workshops, CANDARW 2018  2018年12月 

     詳細を見る

    開催年月日: 2018年11月

    記述言語:英語  

    開催地:Takayama   国名:日本国  

    Recent research has shown that deep convolutional neural networks (DCNN) are vulnerable to several different types of attacks while the reasons of such vulnerability are still under investigation. For instance, the adversarial perturbations can conduct a slight change on a natural image to make the target DCNN make the wrong recognition, while the reasons that DCNN is sensitive to such small modification are divergent from one research to another. In this paper, we evaluate the robustness of two commonly used activation functions of DCNN, namely the sigmoid and ReLu, against the recently proposed low-dimensional one-pixel attack. We show that the choosing of activation functions can be an important factor that influences the robustness of DCNN. The results show that comparing with sigmoid, the ReLu non-linearity is more vulnerable which allows the low dimensional one-pixel attack exploit much higher success rate and confidence of launching the attack. The results give insights on designing new activation functions to enhance the security of DCNN.

  • Lightweight Classification of IoT Malware Based on Image Recognition

    Jiawei Su, Danilo Vasconcellos Vargas, Sanjiva Prasad, Sgandurra Daniele, Yaokai Feng, Kouichi Sakurai

    42nd IEEE Computer Software and Applications Conference, COMPSAC 2018  2018年6月 

     詳細を見る

    開催年月日: 2018年7月

    記述言語:英語  

    開催地:Tokyo   国名:日本国  

    The Internet of Things (IoT) is an extension of the traditional Internet, which allows a very large number of smart devices, such as home appliances, network cameras, sensors and controllers to connect to one another to share information and improve user experiences. IoT devices are micro-computers for domain-specific computations rather than traditional functionspecific embedded devices. This opens the possibility of seeing many kinds of existing attacks, traditionally targeted at the Internet, also directed at IoT devices. As shown by recent events, such as the Mirai and Brickerbot botnets, DDoS attacks have become very common in IoT environments as these lack basic security monitoring and protection mechanisms. In this paper, we propose a novel light-weight approach for detecting DDos malware in IoT environments. We extract the malware images (i.e., a one-channel gray-scale image converted from a malware binary) and utilize a light-weight convolutional neural network for classifying their families. The experimental results show that the proposed system can achieve 94:0% accuracy for the classification of goodware and DDoS malware, and 81:8% accuracy for the classification of goodware and two main malware families.

  • Introducing learning classifier systems Rules that capture complexity

    Ryan J. Urbanowicz, Danilo Vasconcellos Vargas

    2018 Genetic and Evolutionary Computation Conference, GECCO 2018  2018年7月 

     詳細を見る

    開催年月日: 2018年7月

    記述言語:英語  

    開催地:Kyoto   国名:日本国  

  • Effectively Protect Your Privacy Enabling Flexible Privacy Control on Web Tracking

    Shiqian Yu, Danilo Vasconcellos Vargas, Kouichi Sakurai

    5th International Symposium on Computing and Networking, CANDAR 2017  2018年4月 

     詳細を見る

    開催年月日: 2017年11月

    記述言語:英語  

    開催地:Aomori   国名:日本国  

    Third-party tracking, which can collect the users' privacy when users are surfing the Internet, has garnered much attention. Nowadays tracker-blocking tools often use a ruleset based on the domains and elements that need to be blocked. This results in blocking all access tracking, even though the website shows no sign about tracking users' privacy. And what's more, although the tracker-blocking tools try their best to block all the third-party tracking, not all the users dislike the advertisement. Some of them think if their privacy is fine, it's all right to accept advertisements. In this paper, we present a novel framework by using Word2Vec to block third-party tracking. Our goal is to create more flexible and well-developed ruleset that can help users to protect their privacy according to their needs. Instead of blocking all access tracking, we decide to pay more attention to the websites that have a strong probability to collect the users' privacy. We use Word2Vec to classify the websites, and our results show that after using our framework, the error rate drops from 71% to 24%. We believe it brings the new blood into the field of web privacy by providing not only the new third-party tracking tool but also a novel way of thinking about how to block the third-party tracking.

  • Evasion attacks against statistical code obfuscation detectors

    Jiawei Su, Danilo Vasconcellos Vargas, Kouichi Sakurai

    12th International Workshop on Security, IWSEC 2017  2017年1月 

     詳細を見る

    開催年月日: 2017年8月 - 2017年9月

    記述言語:英語  

    開催地:Hiroshima   国名:日本国  

    In the domain of information security, code obfuscation is a feature often employed for malicious purposes. For example there have been quite a few papers reporting that obfuscated JavaScript frequently comes with malicious functionality such as redirecting to external malicious websites. In order to capture such obfuscation, a class of detectors based on statistical features of code, mostly n-grams have been proposed and been claimed to achieve high detection accuracy. In this paper, we formalize a common scenario between defenders who maintain the statistical obfuscation detectors and adversaries who want to evade the detection. Accordingly, we create two kinds of evasion attack methods and evaluate the robustness of statistical detectors under such attacks. Experimental results show that statistical obfuscation detectors can be easily fooled by a sophisticated adversary even in worst case scenarios.

  • Curious Searching for unknown regions of space with a subpopulation-based algorithm

    Danilo Vasconcellos Vargas, Junichi Murata

    2016 Genetic and Evolutionary Computation Conference, GECCO 2016 Companion  2016年7月 

     詳細を見る

    開催年月日: 2016年7月

    記述言語:英語  

    開催地:Denver   国名:アメリカ合衆国  

    Intrinsic motivation and novelty search are promising approaches to deal with plateaus, deceptive functions and other exploration problems where using only the main objective function is insufficient. However, it is not clear until now how and if intrinsic motivation (novelty search) can improve single objective algorithms in general. The hurdle is that using multi-objective algorithms to deal with single-objective problems adds an unnecessary overhead such as the search for non-dominated solutions. Here, we propose the Curious algorithm which is the first multi-objective algorithm focused on solving single-objective problems. Curious uses two subpopulations algorithms. One subpopulation is dedicated for improving objective function values and another one is added to search for unknown regions of space based on objective prediction errors. By using a differential evolution operator, genes from individuals in all subpopulations are mixed. In this way, the promising regions (solutions with high fitness) and unknown regions (solutions with high prediction error) are searched simultaneously. Because of thus realized strong yet well controlled novelty search, the algorithm possesses powerful exploration ability and outperforms usual single population based algorithms such as differential evolution. Thus, it demonstrates that the addition of intrinsic motivation is promising and should improve further single objective algorithms in general.

  • The relationship between (Un)fractured problems and division of input space

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    17th Genetic and Evolutionary Computation Conference, GECCO 2015  2015年7月 

     詳細を見る

    開催年月日: 2015年7月

    記述言語:英語  

    開催地:Madrid   国名:スペイン  

    Problems can be categorized as fractured or unfractured ones. A different set of characteristics are needed for learning algorithms to solve each of these two types of problems. However, the exact characteristics needed to solve each type are unclear. This article shows that the division of the input space is one of these characteristics. In other words, a study is presented showing that while fractured problems benefit from a finer division of the input space, unfractured problems benefit from a coarser division of input space. Many open questions still remains. And the article discusses two conjectures which can be used to solve fractured problems more easily.

  • Novelty-Organizing Team of Classifiers in noisy and dynamic environments

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    IEEE Congress on Evolutionary Computation, CEC 2015  2015年9月 

     詳細を見る

    開催年月日: 2015年5月

    記述言語:英語  

    開催地:Sendai   国名:日本国  

    In the real world, the environment is constantly changing with the input variables under the effect of noise. However, few algorithms were shown to be able to work under those circumstances. Here, Novelty-Organizing Team of Classifiers (NOTC) is applied to the continuous action mountain car as well as two variations of it: A noisy mountain car and an unstable weather mountain car. These problems take respectively noise and change of problem dynamics into account. Moreover, NOTC is compared with NeuroEvolution of Augmenting Topologies (NEAT) in these problems, revealing a trade-off between the approaches. While NOTC achieves the best performance in all of the problems, NEAT needs less trials to converge. It is demonstrated that NOTC achieves better performance because of its division of the input space (creating easier problems). Unfortunately, this division of input space also requires a bit of time to bootstrap.

  • Novelty-organizing team of classifiers - A team-individual multi-objective approach to reinforcement learning

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2014 53rd Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2014  2014年10月 

     詳細を見る

    開催年月日: 2014年9月

    記述言語:英語  

    開催地:Sapporo   国名:日本国  

    In reinforcement learning, there are basically two spaces to search: value-function space and policy space. Consequently, there are two fitness functions each with their associated trade-offs. However, the problem is still perceived as a single-objective one. Here a multi-objective reinforcement learning algorithm is proposed with a structured novelty map population evolving feedforward neural models. It outperforms a gradient based continuous input-output state-of-art algorithm in two problems. Contrary to the gradient based algorithm, the proposed one solves both problems with the same parameters and smaller variance of results. Moreover, the results are comparable even with other discrete action algorithms of the literature as well as neuroevolution methods such as NEAT. The proposed method brings also the novelty map population concept, i.e., a novelty map-based population which is less sensitive to the input distribution and therefore more suitable to create the state space. In fact, the novelty map framework is shown to be less dynamic and more resource efficient than variants of the self-organizing map.

  • Novelty-Organizing Classifiers applied to classification and reinforcement learning Towards flexible algorithms

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    16th Genetic and Evolutionary Computation Conference, GECCO 2014  2014年 

     詳細を見る

    開催年月日: 2014年7月

    記述言語:英語  

    開催地:Vancouver, BC   国名:カナダ  

    It is widely known that reinforcement learning is a more general problem than supervised learning. In fact, supervised learning can be seen as a class of reinforcement learning problems. However, only a couple of papers tested reinforcement learning algorithms in supervised learning problems. Here we propose a new and simpler way to abstract supervised learning for any reinforcement learning algorithm. Moreover, a new algorithm called Novelty-Organizing Classifiers is developed based on a Novelty Map population that focuses more on the novelty of the inputs than their frequency. A comparison of the proposed method with Self-Organizing Classifiers and BioHel on some datasets is presented. Even though BioHel is specialized in solving supervised learning problems, the results showed only a trade-off between the algorithms. Lastly, results on a maze problem validate the flexibility of the proposed algorithm beyond supervised learning problems. Thus, Novelty-Organizing Classifiers is capable of solving many supervised learning problems as well as a maze problem without changing any parameter at all. Considering the fact that no adaptation of parameters was executed, the proposed algorithm's basis seems interestingly flexible.

  • Contingency training

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2013 52nd Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2013  2013年 

     詳細を見る

    開催年月日: 2013年9月

    記述言語:英語  

    開催地:Nagoya   国名:日本国  

    When applied to high-dimensional datasets, feature selection algorithms might still leave dozens of irrelevant variables in the dataset. Therefore, even after feature selection has been applied, classifiers must be prepared to the presence of irrelevant variables. This paper investigates a new training method called Contingency Training which increases the accuracy as well as the robustness against irrelevant attributes. Contingency training is classifier independent. By subsampling and removing information from each sample, it creates a set of constraints. These constraints aid the method to automatically find proper importance weights of the dataset's features. Experiments are conducted with the contingency training applied to neural networks over traditional datasets as well as datasets with additional irrelevant variables. For all of the tests, contingency training surpassed the unmodified training on datasets with irrelevant variables and even outperformed slightly when only a few or no irrelevant variables were present.

  • Continuous adaptive reinforcement learning with the evolution of Self Organizing Classifiers

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2013 IEEE 3rd Joint International Conference on Development and Learning and Epigenetic Robotics, ICDL 2013  2013年 

     詳細を見る

    開催年月日: 2013年8月

    記述言語:英語  

    開催地:Osaka   国名:日本国  

    Learning classifier systems have been solving reinforcement learning problems for some time. However, they face difficulties under multi-step continuous problems. Adaptation may also become harder with time since the convergence of the population decreases its diversity. This article demonstrate that the novel Self Organizing Classifiers method can cope with dynamical multi-step continuous problems. Moreover, adaptation remains the same after convergence.

  • Self organizing classifiers and niched fitness

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2013 15th Genetic and Evolutionary Computation Conference, GECCO 2013  2013年9月 

     詳細を見る

    開催年月日: 2013年7月

    記述言語:英語  

    開催地:Amsterdam   国名:オランダ王国  

    Learning classifier systems are adaptive learning systems which have been widely applied in a multitude of application domains. However, there are still some generalization problems unsolved. The hurdle is that fitness and niching pressures are difficult to balance. Here, a new algorithm called Self Organizing Classifiers is proposed which faces this problem from a different perspective. Instead of balancing the pressures, both pressures are separated and no balance is necessary. In fact, the proposed algorithm possesses a dynamical population structure that self-organizes itself to better project the input space into a map. The niched fitness concept is defined along with its dynamical population structure, both are indispensable for the understanding of the proposed method. Promising results are shown on two continuous multi-step problems. One of which is yet more challenging than previous problems of this class in the literature.

  • A study on the importance of selection pressure and low dimensional weak learners to produce robust ensembles

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    15th Annual Conference on Genetic and Evolutionary Computation, GECCO 2013  2013年 

     詳細を見る

    開催年月日: 2013年7月

    記述言語:英語  

    開催地:Amsterdam   国名:オランダ王国  

    Ensembles of classifiers have been studied for some time. It is widely known that weak learners should be accurate and diverse. However, in the real world there are many constraints and few have been said about the robustness of ensembles and how to develop it. In the context of ran- dom subspace methods, this paper addresses the question of developing ensembles to face problems under time con- straints. Experiments show that selecting weak learners based on their accuracy can be used to create robust en- sembles. Thus, the selection pressure in ensembles is a key technique to create not just effective ensembles but also robust ones. Moreover, the experiments motivate further research on ensembles made of low dimensional classifiers which achieve general accurate results.

  • Multi-objective phylogenetic algorithm Solving multi-objective decomposable deceptive problems

    Jean Paulo Martins, Antonio Helson Mineiro Soares, Danilo Vasconcellos Vargas, Alexandre Cláudio Botazzo Delbem

    6th International Conference on Evolutionary Multi-Criterion Optimization, EMO 2011  2011年4月 

     詳細を見る

    開催年月日: 2011年4月

    記述言語:英語  

    開催地:Ouro Preto   国名:ブラジル連邦共和国  

    In general, Multi-objective Evolutionary Algorithms do not guarantee find solutions in the Pareto-optimal set. We propose a new approach for solving decomposable deceptive multi-objective problems that can find all solutions of the Pareto-optimal set. Basically, the proposed approach starts by decomposing the problem into subproblems and, then, combining the found solutions. The resultant approach is a Multi-objective Estimation of Distribution Algorithm for solving relatively complex multi-objective decomposable problems, using a probabilistic model based on a phylogenetic tree. The results show that, for the tested problem, the algorithm can efficiently find all the solutions of the Pareto-optimal set, with better scaling than the hierarchical Bayesian Optimization Algorithm and other algorithms of the state of art.

  • A New Design for Evaluating Moving Target Defense System

    Wai Kyi Kyi Oo, Hiroshi Koide, Danilo Vasconcellos Vargas, Kouichi Sakurai

    2018年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Empirical Evaluation on Robustness of Deep Convolutional Neural Networks Activation Functions Against Adversarial Perturbation

    Jiawei Su, Danilo Vasconcellos Vargas, Kouichi Sakurai

    2018年7月 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Introducing learning classifier systems: rules that capture complexity

    Ryan J Urbanowicz, Danilo Vasconcellos Vargas

    2018年8月 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Evolutionary reinforcement learning: general models and adaptation.

    Danilo Vasconcellos Vargas

    2018年8月 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Connection-Aware Spectrum-Diversity for Neuroevolution

    Danilo Vasconcellos VARGAS, Yuta INOUE

    2019年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Neural Cryptography Based on the Topology Evolving Neural Networks

    Yuetong Zhu, Danilo Vasconcellos Vargas, Kouichi Sakurai

    2018年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Continuous adaptive reinforcement learning with the evolution of self organizing classifiers

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2013年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • A study on the importance of selection pressure and low dimensional weak learners to produce robust ensembles

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2013年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Novelty-organizing team of classifiers-a team-individual multi-objective approach to reinforcement learning

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2014年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Novelty-organizing classifiers applied to classification and reinforcement learning: towards flexible algorithms

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2014年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • The Relationship Between (Un) Fractured Problems and Division of Input Space

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2015年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Novelty-organizing team of classifiers in noisy and dynamic environments

    Danilo Vasconcellos Vargas, Hirotaka Takano, Junichi Murata

    2015年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Curious: Searching for Unknown Regions of Space with a Subpopulation-based Algorithm

    Danilo Vasconcellos Vargas, Junichi Murata

    2016年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Effectively Protect Your Privacy: Enabling Flexible Privacy Control on Web Tracking

    Shiqian Yu, Danilo Vasconcellos Vargas, Kouichi Sakurai

    2017年 

     詳細を見る

    記述言語:その他  

    国名:その他  

  • Tracing MIRAI Malware in Networked System

    Yao Xu, Hiroshi Koide, Danilo Vasconcellos Vargas, Kouichi Sakurai

    2018年 

     詳細を見る

    記述言語:その他  

    国名:その他  

▼全件表示

所属学協会

  • IEEE

学術貢献活動

  • 委員 国際学術貢献

    AAAI 2024 (Core Rank: A*, One of the Top AI Conferences)  ( その他 ) 2024年2月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    GECCO 2024 (ACM, Core Rank: A)  ( その他 ) 2023年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    AAAI 2023 (Core Rank: A*, One of the Top AI Conferences)  ( その他 ) 2023年2月

     詳細を見る

    種別:大会・シンポジウム等 

  • 学術論文等の審査

    役割:査読

    2023年

     詳細を見る

    種別:査読等 

    外国語雑誌 査読論文数:10

    日本語雑誌 査読論文数:0

    国際会議録 査読論文数:8

    国内会議録 査読論文数:0

  • 委員 国際学術貢献

    GECCO 2022 (ACM, Core Rank: A)  ( その他 ) 2022年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    AAAI 2022 (Core Rank: A*, One of the Top AI Conferences)  ( その他 ) 2022年2月

     詳細を見る

    種別:大会・シンポジウム等 

  • 学術論文等の審査

    役割:査読

    2022年

     詳細を見る

    種別:査読等 

    外国語雑誌 査読論文数:10

    日本語雑誌 査読論文数:0

    国際会議録 査読論文数:10

    国内会議録 査読論文数:0

  • 委員 国際学術貢献

    GECCO 2021 (ACM, Core Rank: A)  ( その他 ) 2021年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    Twenty-Second International Workshop on Learning Classifier Systems  ( その他 ) 2021年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 学術論文等の審査

    役割:査読

    2021年

     詳細を見る

    種別:査読等 

    外国語雑誌 査読論文数:20

    日本語雑誌 査読論文数:0

    国際会議録 査読論文数:20

    国内会議録 査読論文数:0

  • 委員 国際学術貢献

    GECCO 2020 (ACM, Core Rank: A)  ( その他 ) 2020年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    Twenty-Second International Workshop on Learning Classifier Systems  ( その他 ) 2020年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    IEEE CEC 2020  ( その他 ) 2020年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 学術論文等の審査

    役割:査読

    2020年

     詳細を見る

    種別:査読等 

    外国語雑誌 査読論文数:8

    日本語雑誌 査読論文数:0

    国際会議録 査読論文数:10

    国内会議録 査読論文数:0

  • 委員 国際学術貢献

    GECCO 2019 (ACM, Core Rank: A)  ( その他 ) 2019年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    Twenty-Second International Workshop on Learning Classifier Systems  ( その他 ) 2019年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    IEEE CEC 2019  ( その他 ) 2019年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 学術論文等の審査

    役割:査読

    2019年

     詳細を見る

    種別:査読等 

    外国語雑誌 査読論文数:5

    日本語雑誌 査読論文数:0

    国際会議録 査読論文数:12

    国内会議録 査読論文数:0

  • 幹事 国際学術貢献

    Twenty-First International Workshop on Learning Classifier Systems (at GECCO 2018)  ( 京都 ) 2018年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    GECCO 2018 (ACM, Core Rank: A)  ( その他 ) 2018年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 委員 国際学術貢献

    IEEE CEC 2018  ( その他 ) 2018年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • 学術論文等の審査

    役割:査読

    2018年

     詳細を見る

    種別:査読等 

    外国語雑誌 査読論文数:0

    日本語雑誌 査読論文数:0

    国際会議録 査読論文数:12

    国内会議録 査読論文数:0

  • Organizer 国際学術貢献

    Evolutionary Rule-based Machine Learning Workshop (Twentieth International Workshop on Learning Classifier Systems)  ( Berlin Germany ) 2017年7月

     詳細を見る

    種別:大会・シンポジウム等 

    参加者数:50

  • 委員 国際学術貢献

    IEEE CEC 2017  ( その他 ) 2017年7月

     詳細を見る

    種別:大会・シンポジウム等 

  • Organizer 国際学術貢献

    Workshop on Cybersecurity for IoT - Towards Secure Smart Buildings  ( Fukuoka Japan ) 2017年7月

     詳細を見る

    種別:大会・シンポジウム等 

    参加者数:30

  • 学術論文等の審査

    役割:査読

    2017年

     詳細を見る

    種別:査読等 

    外国語雑誌 査読論文数:2

    日本語雑誌 査読論文数:0

    国際会議録 査読論文数:5

    国内会議録 査読論文数:0

▼全件表示

共同研究・競争的資金等の研究課題

  • 光の極限性能を生かすフォトニックコンピューティングの創成

    2023年4月 - 2027年3月

    学術変革領域研究 (A) 

      詳細を見る

    担当区分:研究分担者 

  • 光の極限性能を生かすフォトニックコンピューティングの創成

    研究課題/領域番号:22H05194  2023年 - 2027年

    日本学術振興会・文部科学省  科学研究費助成事業  学術変革領域研究(A)

      詳細を見る

    担当区分:研究分担者  資金種別:科研費

  • 使途特定寄附金

    2023年

      詳細を見る

    資金種別:寄附金

  • Small Anomaly Detection model development-small anomaly detection algorithm for autonomous driving based on Inpainting

    2022年12月 - 2024年12月

    Huawei 

      詳細を見る

    担当区分:研究代表者 

  • Small Anomaly Detection model development-small anomaly detection algorithm for autonomous driving based on lnpainting

    2022年12月 - 2023年12月

    受託研究

      詳細を見る

    担当区分:研究代表者  資金種別:その他産学連携による資金

  • 自己組織化による学習:人工知能の新たな基礎を開拓

    2022年8月 - 2024年4月

    挑戦的研究(萌芽) 

      詳細を見る

    担当区分:研究代表者 

  • Data Fusion based Hybrid Deep Learning

    2022年8月 - 2024年4月

    JST - Acceleration Phase – Program AIP 

      詳細を見る

    担当区分:研究分担者 

  • 自己組織化による学習:人工知能の新たな基礎を開拓

    研究課題/領域番号:22K19814  2022年 - 2023年

    日本学術振興会  科学研究費助成事業  挑戦的研究(萌芽)

      詳細を見る

    担当区分:研究代表者  資金種別:科研費

  • デジタルツインを用いた自動運転AIの検証と妥当性確認

    2022年 - 2023年

    科学技術振興費(主要5分野) (文部科学省)

      詳細を見る

    担当区分:研究分担者  資金種別:受託研究

  • Research of Robust and Domain-invariant Algorithm for Machine Vision and Image Processing

    2021年10月 - 2023年10月

    受託研究

      詳細を見る

    担当区分:研究代表者  資金種別:その他産学連携による資金

  • Research of Robust and Domain-invariant Algorithm for Machine Vision and Image Processing

    2021年10月 - 2023年10月

    Huawei 

      詳細を見る

    担当区分:研究代表者 

  • Learning Internal Representations Robust against Adversarial Attacks

    2020年4月 - 2022年4月

    日本学術振興会 科学研究費助成事業 若手研究 

      詳細を見る

    担当区分:研究代表者 

  • Learning Internal Representations Robust against Adversarial Attacks

    研究課題/領域番号:20K19824  2020年 - 2021年

    日本学術振興会  科学研究費助成事業  若手研究

      詳細を見る

    担当区分:研究代表者  資金種別:科研費

  • 頑強なハイブリッド深層学習モデルの自動探索システム (加速フェーズ)

    2020年 - 2021年

    科学技術振興調整費 (文部科学省)

      詳細を見る

    担当区分:研究代表者  資金種別:受託研究

  • ディープラーニングとニューロエボリューション(SUNA)を用いた教師データあり学習の船舶機械異常検知への適用

    2019年2月 - 2019年7月

    共同研究

      詳細を見る

    担当区分:研究代表者  資金種別:その他産学連携による資金

  • 頑強なハイブリッド深層学習モデルの自動探索システム

    2018年 - 2019年

    科学技術振興調整費 (文部科学省)

      詳細を見る

    担当区分:研究代表者  資金種別:受託研究

  • Misleading Algorithms: Interdisciplinary Perspectives on the Implications for Law

    2018年 - 2019年

    九州大学 QR Tsubasa

      詳細を見る

    担当区分:研究分担者  資金種別:学内資金・基金等

  • Learning to Run with High Dimensional Robots – A Reinforcement Learning Approach

    2017年

    九州大学 わかばチャレンジ (整理番号:29254)

      詳細を見る

    担当区分:研究代表者  資金種別:学内資金・基金等

  • 攻撃によって学習システムを解析する研究

    2017年

    九州大学 データサイエンス実践教育経費公募分

      詳細を見る

    担当区分:研究代表者  資金種別:学内資金・基金等

  • 次世代機械学習の開発

    2016年

    九州大学 スタートアップ支援経費

      詳細を見る

    担当区分:研究代表者  資金種別:学内資金・基金等

▼全件表示

担当授業科目

  • (IUPE)Programming MethodologyⅠ

    2024年4月 - 2024年9月   前期

  • C課程実験

    2024年4月 - 2024年9月   前期

  • AI and Law

    2023年6月 - 2023年8月   夏学期

  • (IUPE)Programming PracticeⅠ

    2023年4月 - 2023年9月   前期

  • (IUPE)Programming MethodologyⅠ

    2023年4月 - 2023年9月   前期

  • (IUPE)Programming PracticeⅠ

    2023年4月 - 2023年9月   前期

  • C課程実験

    2023年4月 - 2023年9月   前期

  • (IUPE)Programming MethodologyⅠ

    2022年4月 - 2022年6月   春学期

  • (IUPE)Programming PracticeⅠ

    2022年4月 - 2022年6月   春学期

  • 情報科学

    2021年10月 - 2022年3月   後期

  • 情報科学(英語)

    2021年10月 - 2022年3月   後期

  • 情報理工学演示

    2021年10月 - 2022年3月   後期

  • 国際科学特論Ⅱ

    2021年10月 - 2021年12月   秋学期

  • (IUPE)Fund. of Electrical Eng and Computer Science I

    2021年10月 - 2021年12月   秋学期

  • Seminar in Information Science and Technology

    2021年4月 - 2022年3月   通年

  • 情報理工学研究Ⅰ

    2021年4月 - 2022年3月   通年

  • 情報理工学演習

    2021年4月 - 2022年3月   通年

  • Research in Information Science and Technology I

    2021年4月 - 2022年3月   通年

  • (IUPE) Programming Methodology Ⅰ

    2021年4月 - 2021年9月   前期

  • 情報理工学読解

    2021年4月 - 2021年9月   前期

  • [M2]情報学論述Ⅰ

    2021年4月 - 2021年9月   前期

  • [M2]情報学論議Ⅰ

    2021年4月 - 2021年9月   前期

  • Technical Reading in Information Science and Technology

    2021年4月 - 2021年9月   前期

  • [M2]情報学読解

    2021年4月 - 2021年9月   前期

  • [M2]情報学演示

    2021年4月 - 2021年9月   前期

  • C課程実験

    2021年4月 - 2021年9月   前期

  • (IUPE)Programming Practice I

    2021年4月 - 2021年9月   前期

  • (IUPE)Programming PracticeⅠ

    2021年4月 - 2021年6月   春学期

  • (IUPE)Programming MethodologyⅠ

    2021年4月 - 2021年6月   春学期

  • 情報科学

    2020年10月 - 2021年3月   後期

  • 情報学演示

    2020年10月 - 2021年3月   後期

  • 情報学論述Ⅱ

    2020年10月 - 2021年3月   後期

  • 情報学論議Ⅱ

    2020年10月 - 2021年3月   後期

  • 国際科学特論Ⅱ

    2020年10月 - 2020年12月   秋学期

  • 情報論理学

    2020年4月 - 2020年9月   前期

  • 情報学論述Ⅰ

    2020年4月 - 2020年9月   前期

  • 情報学論議Ⅰ

    2020年4月 - 2020年9月   前期

  • 情報論理学

    2020年4月 - 2020年9月   前期

  • 情報科学

    2019年10月 - 2020年3月   後期

  • 情報科学

    2019年10月 - 2020年3月   後期

  • 人工知能:ニューラルネットワークと進化計算

    2019年6月 - 2019年8月   夏学期

  • 暗号と情報セキュリティ

    2019年4月 - 2019年9月   前期

  • 人工知能:ニューラルネットワークと進化計算

    2019年4月 - 2019年6月   春学期

  • 情報科学

    2018年10月 - 2019年3月   後期

  • 人工知能Ⅱ:最先端を理解する

    2018年6月 - 2018年8月   夏学期

  • 暗号と情報セキュリティ

    2018年4月 - 2018年9月   前期

  • 人工知能Ⅰ:基礎を理解する

    2018年4月 - 2018年6月   春学期

  • 暗号と情報セキュリティ

    2017年4月 - 2017年9月   前期

  • 高度プログラミング

    2017年4月 - 2017年9月   前期

  • 数値解析演習

    2017年4月 - 2017年9月   前期

  • ハードウェア実験

    2017年4月 - 2017年9月   前期

▼全件表示

FD参加状況

  • 2023年10月   役割:参加   名称:【シス情FD】価値創造型半導体人材育成センターについて

    主催組織:部局

  • 2023年9月   役割:参加   名称:【シス情FD】Top10%論文/Top10%ジャーナルとは何か: 傾向と対策

    主催組織:部局

  • 2023年4月   役割:参加   名称:【シス情FD】若手教員による研究紹介⑧

    主催組織:部局

  • 2023年1月   役割:参加   名称:【シス情FD】若手教員による研究紹介⑦

    主催組織:部局

  • 2022年11月   役割:参加   名称:【工学・シス情】教職員向け知的財産セミナー(FD)

    主催組織:部局

  • 2022年4月   役割:参加   名称:【シス情FD】第4期中期目標・中期計画等について

    主催組織:部局

  • 2022年1月   役割:参加   名称:【シス情FD】シス情関連の科学技術に対する国の政策動向(に関する私見)

    主催組織:部局

  • 2021年9月   役割:参加   名称:JST 次世代研究者挑戦的研究プログラム 説明会

    主催組織:全学

  • 2021年9月   役割:参加   名称:博士後期課程の充足率向上に向けて

    主催組織:部局

  • 2021年5月   役割:参加   名称:先導的人材育成フェローシップ事業(情報・AI分野)について

    主催組織:部局

  • 2021年3月   役割:参加   名称:オンライン授業におけるアクティブラーニングの科目の実施状況:アクティブラーニングの今後を考える

    主催組織:部局

  • 2021年3月   役割:参加   名称:FD講演会「九州大学オンライン授業のグッドプラクティス 〜 オンデマンド型授業編〜」

    主催組織:部局

  • 2020年11月   役割:参加   名称:マス・フォア・イノベーション卓越大学院について

    主催組織:部局

  • 2020年10月   役割:参加   名称:2020年度 ユニバーシティ・デザイン・ワークショップの報告

    主催組織:部局

  • 2020年9月   役割:参加   名称:電気情報工学科総合型選抜(AO入試)について

    主催組織:部局

  • 2020年7月   役割:参加   名称:アフターコロナの大学はどうあるべきか

    主催組織:部局

  • 2020年4月   役割:参加   名称:新型コロナウイルスが誘起した社会変化に対する システム情報科学からの提言

    主催組織:全学

  • 2020年2月   役割:参加   名称:九州大学工学系改組の現状と今後の予定

    主催組織:全学

  • 2019年7月   役割:参加   名称:3ポリシーに関する全学FD ~日本学術会議分野別参照基準に基づく理学部物理学科の3ポリシー~

    主催組織:全学

  • 2019年4月   役割:参加   名称:3ポリシーの見直し指針

    主催組織:全学

  • 2019年3月   役割:参加   名称:M2B学習支援システム講習会

    主催組織:全学

  • 2018年5月   役割:参加   名称:LIFE

    主催組織:全学

  • 2018年1月   役割:参加   名称:Setting Up International Collaborations

    主催組織:全学

  • 2017年12月   役割:参加   名称:自殺防止メンタルヘルス研修会

    主催組織:全学

  • 2017年12月   役割:参加   名称:大学生の学習時間

    主催組織:全学

  • 2017年10月   役割:参加   名称:いよいよスタートした電気情報工学科国際コース

    主催組織:全学

  • 2017年3月   役割:参加   名称:外務省「インド情報技術大学ジャバルプール校(IIIT-DMJ)のための日印協力」 による日本人教員派遣プログラム参加報告

    主催組織:全学

▼全件表示

他大学・他機関等の客員・兼任・非常勤講師等

  • 2024年  東京大学  区分:客員教員 

  • 2023年  東京大学  区分:客員教員 

  • 2022年  東京大学  区分:客員教員 

  • 2021年  東京大学  区分:客員教員 

  • 2020年  東京大学  区分:客員教員 

その他教育活動及び特記事項

  • 2023年  クラス担任  学部

  • 2021年  クラス担任  学部

  • 2020年  学友会・同好会等の指導  Open Lab (AI Jam + Game Jam)

     詳細を見る

    創設者

  • 2020年  その他特記事項  Adversarial Machine Learning: On The Deeper Secrets of Deep Learning 学会:WCCI 2020 (トップ学会のチュートリアルの講演を行う)

     詳細を見る

    Adversarial Machine Learning: On The Deeper Secrets of Deep Learning
    学会:WCCI 2020
    (トップ学会のチュートリアルの講演を行う)

  • 2019年  その他特記事項  Adversarial Machine Learning: On The Deeper Secrets of Deep Learning 学会:IJCAI 2020 (トップ学会のチュートリアルの講演を行う)

     詳細を見る

    Adversarial Machine Learning: On The Deeper Secrets of Deep Learning
    学会:IJCAI 2020
    (トップ学会のチュートリアルの講演を行う)

  • 2018年  その他特記事項  Learning Classifier Systemsの導入: 複雑を表現できるルール 学会:GECCO 2018 (トップ学会のチュートリアルの講演を行う)

     詳細を見る

    Learning Classifier Systemsの導入: 複雑を表現できるルール
    学会:GECCO 2018
    (トップ学会のチュートリアルの講演を行う)

  • 2018年  その他特記事項  進化計算に基づく強化学習: 一般的な学習モデル と適応 学会:GECCO 2018 (トップ学会のチュートリアルの講演を行う)

     詳細を見る

    進化計算に基づく強化学習: 一般的な学習モデル と適応
    学会:GECCO 2018
    (トップ学会のチュートリアルの講演を行う)

  • 2017年  学友会・同好会等の指導  AI-Q - 人工知能に関するサークル

     詳細を見る

    創設者

▼全件表示

社会貢献・国際連携活動概要

  • AIの研究開発への関心を高めるために、自律型ドローンのコンテストを企画しています。
    SDGsに取り組む企業と、あるいは企業内で複数のプロジェクトを立ち上げる予定です。
    私が最近設立した会社は、少なくとも 5 つの SDG に直接取り組んでおり、より多くの社会経済的影響と健全な起業家精神のエコシステムの構築に貢献したいと考えています。

社会貢献活動

  • 松田先生と行っているコラボの研究成果をビジネス化ができるため、PARKS(という九州のstart upのプログラム)に応募し、活動しています

    福岡市、九州、OIP  福岡  2024年3月

     詳細を見る

    対象:社会人・一般, 学術団体, 企業, 市民団体, 行政機関

    種別:その他

  • International Schoolの改善を目指せ、福岡市と九州大学のOIPと一緒に教育プログラムを提案し、議論を進めています。

    福岡市、OIP、Fukuoka International School  福岡  2024年3月

     詳細を見る

    対象:社会人・一般, 学術団体, 企業, 市民団体, 行政機関

    種別:その他

    トップの外国人教員・スタッフをキープするため、より良いインターナショナルスクールづくりに挑戦する。

  • 人工知能がもたらす九州の未来社会

    九州大学  JR博多シティ会議室  2022年11月

     詳細を見る

    対象:社会人・一般, 学術団体, 企業, 市民団体, 行政機関

    種別:講演会

  • 人工知能ってなに?どこにある?

    九州大学のオープンキャンパス  九州大学  2019年8月

     詳細を見る

    対象:社会人・一般, 学術団体, 企業, 市民団体, 行政機関

    種別:講演会

メディア報道

  • One-pixel Attackに関する記載 新聞・雑誌

    AI Magazine (ISSN 0738-4602)  2021年4月

     詳細を見る

    One-pixel Attackに関する記載

  • "AI image recognition fooled by single pixel change" 新聞・雑誌

    BBC News  2017年10月

     詳細を見る

    "AI image recognition fooled by single pixel change"

政策形成、学術振興等への寄与活動

  • 2023年6月 - 2024年6月   GECCO (Core rank A)

    研究委員会委員

  • 2023年6月 - 2024年6月   AAAI

    研究委員会委員

  • 2022年6月 - 2023年6月   AAAI

    研究委員会委員

  • 2021年6月 - 2022年6月   GECCO (Core rank A)

    研究委員会委員

  • 2021年6月 - 2022年6月   AAAI

    研究委員会委員

  • 2020年6月 - 2021年6月   IJCAI

    研究委員会委員

  • 2019年6月 - 2021年6月   GECCO (Core rank A)

    研究委員会委員

  • 2019年6月 - 2021年6月   CEC (Core Rank B)

    研究委員会委員

  • 2018年6月   CECのトップIEEE学会

    研究委員会委員

  • 2018年6月   GECCOのトップACM学会

    研究委員会委員

  • 2017年6月 - 2021年6月   Twenty-Second International Workshop on Learning Classifier Systems

    研究委員会委員

▼全件表示

海外渡航歴

  • 2023年6月

    滞在国名1:スウェーデン王国   滞在機関名1:University of Linkoping

  • 2022年6月

    滞在国名1:ブラジル連邦共和国   滞在機関名1:University of Sao Paulo

  • 2018年9月

    滞在国名1:インド   滞在機関名1:Indian Institute of Technology Delhi

  • 2017年8月

    滞在国名1:インド   滞在機関名1:Indian Institute of Technology Delhi

学内運営に関わる各種委員・役職等

  • 2024年5月 - 2024年10月   研究院 グローバルコース修論試問の幹事

  • 2024年4月 - 2027年3月   研究院 TAとりまとめ担当教員

  • 2022年4月 - 2023年3月   その他 国際卓越研究大学に係る研究ワーキンググループの委員

  • 2022年4月 - 2023年3月   その他 国際卓越研究大学に係る国際ワーキンググループの委員

  • 2022年4月 - 2023年3月   研究院 次世挑 SPRING - 情報系代表

  • 2021年4月 - 2023年3月   その他 国際コースワーキンググループの委員

  • 2021年4月 - 2022年3月   その他 オープンキャンパスワーキンググループの委員

  • その他 東京大学 大学院工学系研究科 役員研究員

▼全件表示