Eiji Takimoto | Last modified date：2022.11.08 |

Professor /
Intelligence Science /
Department of Informatics /
Faculty of Information Science and Electrical Engineering

**Papers**

1. | Daiki Suehiro, Eiji Takimoto, Simplified and unified analysis of various learning problems by reduction to Multiple-Instance Learning, Proceedings of the 38th Conference on Uncertainty in Artificial Intelligence (UAI 2022), PMLR, 180, 1896-1906, 2022.08. |

2. | Yaxiong Liu, Ken-ichiro Moridomi, Kohei Hatano, Eiji Takimoto, An Online Semi-definite Programming with a Generalized Log-determinant Regularizer and Its Applications, Mathematics, 10.3390/math10071055, 10, 7, 1-22, 2022.03, [URL]. |

3. | Sherief Hashima, Ehab Mahmoud Mohamed, Kohei Hatano, Eiji Takimoto, WiGig Wireless Sensor Selection Using Sophisticated Multi Armed Bandit Schemes, Proceedings of 13th International Conference on Mobile Computing and Ubiquitous Network (ICMU 2021), 10.23919/ICMU50196.2021.9638849, 1-6, 2021.11. |

4. | Yaxiong Liu, Ken-ichiro Moridomi, Kohei Hatano, Eiji Takimoto, An Online Semi-definite Programming With a Generalised Log-determinant Regularizer And Its Applications, Proceedings of 13th Asian Conference on Machine Learning (ACML 2021), PMLR, 157, 1113-1128, 2021.11. |

5. | Yaxiong Liu, Xuanke Jiang, Kohei Hatano, Eiji Takimoto, Expert Advice Problem With Noisy Low Rank Loss, Proceedings of 13th Asian Conference on Machine Learning (ACML 2021), PMLR, 157, 1097-1112, 2021.11. |

6. | Liu Yaxiong, Kohei Hatano, Eiji Takimoto, Improved Algorithms for Online Load Balancing, Proceedings of the 47th International Conference on Current Trends in Theory and Practice of Computer Science (SOFSEM 2021), LNCS, 10.1007/978-3-030-67731-2_15, 12607, 203-217, 2021.01. |

7. | Sherief Hashima, Kohei Hatano, Eiji Takimoto, Ehab Mahmoud Mohamed, Minimax Optimal Stochastic Strategy (MOSS) For Neighbor Discovery and Selection in Millimeter Wave D2D Networks, Proceedings of the 23rd International Symposium on Wireless Personal Multimedia Communications (WPMC 2020), 10.1109/WPMC50192.2020.9309495, 1-6, 2020.10. |

8. | Daiki Suehiro, Kohei Hatano, Eiji Takimoto, Shuji Yamamoto, Kenichi Bannai, Akiko Takeda, Theory and Algorithms for Shapelet-Based Multiple-Instance Learning, Neural Computation, 10.1162/neco_a_01297, 32, 8, 1580-1613, 2020.08, We propose a new formulation of multiple-instance learning (MIL), in which a unit of data consists of a set of instances called a bag. The goal is to find a good classifier of bags based on the similarity with a "shapelet" (or pattern), where the similarity of a bag with a shapelet is the maximum similarity of instances in the bag. In previous work, some of the training instances have been chosen as shapelets with no theoretical justification. In our formulation, we use all possible, and thus infinitely many, shapelets, resulting in a richer class of classifiers. We show that the formulation is tractable, that is, it can be reduced through linear programming boosting (LPBoost) to difference of convex (DC) programs of finite (actually polynomial) size. Our theoretical result also gives justification to the heuristics of some previous work. The time complexity of the proposed algorithm highly depends on the size of the set of all instances in the training sample. To apply to the data containing a large number of instances, we also propose a heuristic option of the algorithm without the loss of the theoretical guarantee. Our empirical study demonstrates that our algorithm uniformly works for shapelet learning tasks on time-series classification and various MIL tasks with comparable accuracy to the existing methods. Moreover, we show that the proposed heuristics allow us to achieve the result in reasonable computational time.. |

9. | Sherief Hashima, Kohei Hatano, Eiji Takimoto, Ehab Mahmoud Mohamed, Neighbor Discovery and Selection in Millimeter Wave D2D Networks Using Stochastic MAB, IEEE Communications Letters, 10.1109/LCOMM.2020.2991535, 24, 8, 1840-1844, 2020.08, The propagation characteristics of millimeter-wave (mmWaves), encourages its use in the device to device (D2D) communications for fifth-generation (5G) and future beyond 5G (B5G) networks. However, due to the use of beamforming training (BT), there is a tradeoff between exploring neighbor devices for best device selection and the required overhead. In this letter, using a tool of machine learning, joint neighbor discovery and selection (NDS) in mmWave D2D networks is formulated as a stochastic budget-constraint multi-armed bandit (MAB) problem. Hence, a modified Thomson sampling (TS) and variants of upper confidence bound (UCB) based algorithms are proposed to address the topic while considering the residual energies of the surrounding devices. Simulation analysis demonstrates the effectiveness of the proposed techniques over the conventional approaches concerning average throughput, energy efficiency, and network lifetime.. |

10. | Takahiro Fujita, Kohei Hatano, Eiji Takimoto, Boosting over non-deterministic ZDDs, Theoretical Computer Science, 10.1016/j.tcs.2018.11.027, 806, 81-89, 2020.02. |

11. | Yiping Tang, Kohei Hatano, Eiji Takimoto, Recognition of Japanese Historical Hand-Written Characters Based on Object Detection Methods, Proceedings of the 5th International Workshop on Historical Document Imaging and Processing,
HIP@ICDAR 2019, 10.1145/3352631.3352642, 72-77, 2019.09. |

12. | Fumito Miyake, Eiji Takimoto, Kohei Hatano, Succinct Representation of Linear Extensions via MDDs and Its Application to Scheduling Under Precedence Constraints, Proceedings of the 30th International Workshop on Combinatorial Algorithms (IWOCA 2019), 10.1007/978-3-030-25005-8_30, 11638, 365-377, 2019.07. |

13. | Takahiro Fujita, Kohei Hatano, Shuji Kijima, Eiji Takimoto, Online Combinatorial Optimization with Multiple Projections and Its Application to Scheduling Problem, IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 10.1587/transfun.E101.A.1334, E101.A, 9, 1334-1343, 2018.09. |

14. | Ken-ichiro Moridomi, Kohei Hatano, Eiji Takimoto, Tighter Generalization Bounds for Matrix Completion Via Factorization Into Constrained Matrices, IEICE Transactions on Information and Systems, 10.1587/transinf.2017EDP7339, E101.D, 8, 1997-2004, 2018.08. |

15. | Kosuke Matsumoto, Kohei Hatano, Eiji Takimoto, Decision Diagrams for Solving a Job Scheduling Problem Under Precedence Constraints, Proceedings of the 17th International Symposium on Experimental Algorithms (SEA 2018), 10.4230/LIPIcs.SEA.2018.5, 5:1-5:12, 2018.06. |

16. | Ken-ichiro Moridomi, Kohei Hatano, Eiji Takimoto, Online Linear Optimization with the Log-Determinant Regularizer, IEICE Transactions on Information and Systems, 10.1587/transinf.2017EDP7317, E101-D, 6, 1511-1520, 2018.06. |

17. | Daiki Suehiro, Kohei Hatano, Eiji Takimoto, Efficient Reformulation of 1-Norm Ranking SVM, IEICE Transactions on Information and Systems, E101-D, 3, 719-729, 2018.03. |

18. | Takahiro Fujita, Kohei Hatano, Eiji Takimoto, Boosting over Non-deterministic ZDDs, Proceedings of the 12th International Conference and Workshops on Algorithms and Computation (WALCOM 2018), 10.1007/978-3-319-75172-6_17, 10755, 195-206, Lecture Notes in Computer Science, 2018.03. |

19. | Takumi Nakazono, Ken-ichiro Moridomi, Kohei Hatano, Eiji Takimoto, A Combinatorial Metrical Task System Problem Under the Uniform Metric, Proceedings of the 27th International Conference on Algorithmic Learning Theory (ALT 2017), 10.1007/978-3-319-46379-7_19, 9925, 1577-1586, 2016.10. |

20. | Nir Ailon, Kohei Hatano, Eiji Takimoto, Bandit Online Optimization Over the Permutahedron , Theoretical Computer Science, 10.1016/j.tcs.2016.07.033, 650, 92-108, 2016.10. |

21. | Takahiro Fujita, Kohei Hatano, Shuji Kijima, Eiji Takimoto, Online Linear Optimization for Job Scheduling under Precedence Constraints, Proc. 26th International Conference on Algorithmic Learning Theory (ALT 2015), 10.1007/978-3-319-24486-0_22, 9355, 332-346, 2015.10. |

22. | Peter Bartlett, Wouter Koolen, Alan Malek, Eiji Takimoto, Manfred Warmuth, Minimax Fixed-Design Linear Regression, Proceedings of The 28th Conference on Learning Theory (COLT 2015), 40, 226-239, 2015.07. |

23. | Issei Matsumoto, Kohei Hatano, Eiji Takimoto, Online Density Estimation of Bradley-Terry Models, Proceedings of The 28th Conference on Learning Theory (COLT 2015)
, 40, 1343-1359, 2015.07. |

24. | Kei Uchizawa, Eiji Takimoto, Lower bounds for linear decision trees with bounded weights, Proc. 41st International Conference on Current Trends in Theory and Practice of Computer Science (SOFSEM 2015), 10.1007/978-3-662-46078-8_34, 8939, 412-422, 2015.01. |

25. | Ken-ichiro Moridomi, Kohei Hatano, Eiji Takimoto, Online matrix prediction for sparse loss matrices, Proc. 6th Asian Conference on Machine Learning (ACML 2014), 39, 250-265, 2015.02. |

26. | Nir Ailon, Kohei Hatano, Eiji Takimoto, Bandit online optimization over permutahedron, Proc. 25th International Conference on Algorithmic Learning Theory (ALT 2014)
, 10.1007/978-3-319-11662-4_16, 8776, 215-229, 2014.10. |

27. | Kazuki Teraoka, Kohei Hatano, Eiji Takimoto, Efficient Sampling Method for Monte Carlo Tree Search, IEICE Transactions on Information and Systems, E97-D, 3, 392-398, 2014.03. |

28. | Takahiro Fujita, Kohei Hatano, Eiji Takimoto, Combinatorial Online Prediction via Metarounding, 24th International Conference on Algorithmic Learning Theory (ALT 2013), 8139, 68-82, 2013.10. |

29. | Eiji Takimoto, Kohei Hatano, Efficient Algorithms for Combinatorial Online Prediction, 24th International Conference on Algorithmic Learning Theory (ALT 2013), 8139, 22-32, 2013.10. |

30. | Shota Yasutake, Kohei Hatano, Eiji Takimoto, Masayuki Takeda, Online Rank Aggregation, 4th Asian Conference on Machine Learning (ACML 2012), 25, 539-553, 2012.11. |

31. | Daiki Suehiro, Kohei Hatano, Shuji Kijima, Kiyohito Nagano, Eiji Takimoto, Online Prediction under Submodular Constraints, 23rd International Conference on Algorithmic Learning Theory (ALT 2012), 7568, 260-274, 2012.10. |

32. | Shota Yasutake, Kohei Hatano, Shuji Kijima, Eiji Takimoto, Masayuki Takeda, Online Linear Optimization over Permutations, 22nd International Symposium on Algorithms and Computation (ISAAC 2011), LNCS, 7074, 534-543, 2011.12. |

33. | Shin-ichi Yoshida, Kohei Hatano, Eiji Takimoto, Masayuki Takeda, Adaptive Online Prediction Using Weighted Windows, IEICE Transactions on Information and Systems, E94-D, 10, 1917-1923, 2011.10. |

34. | Daiki Suehiro, Kohei Hatano, Eiji Takimoto, Approximate Reduction from AUC Maximization to 1-norm Soft Margin Optimization, 22nd International Conference on Algorithmic Learning Theory (ALT 2011), LNAI, 6925, 324-337, 2011.10. |

35. | Kei Uchizawa, Eiji Takimoto, Lower Bounds for Linear Decision trees via An Energy Complexity Argument, 36th International Symposiums on Mathematical Foundations of Computer Science (MFCS 2011), LNCS, 6907, 568-579, 2011.08. |

36. | Kei Uchizawa, Eiji Takimoto, Takao Nishizeki , Size–energy tradeoffs for unate circuits computing symmetric Boolean functions, Theoretical Computer Science, 10.1016/j.tcs.2010.11.022, 412, 773-782, 2011.03. |

37. | Kei Uchizawa, Takao Nishizeki, Eiji Takimoto, Energy and depth of threshold circuits, Theoretical Computer Science, 10.1016/j.tcs.2010.08.006, 411, 3938-3946, 2010.10. |

38. | Hideaki Fukuhara, Eiji Takimoto, Kazuyuki Amano, NPN-representatives of a Set of Optimal Boolean Formulas, IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, E-93A, 6, 2010.06. |

39. | Hideaki Fukuhara, Eiji Takimoto, Lower Bounds on Quantum Query Complexity for Read-once Formulas with XOR and MUX Operators, IEICE Transactions on Information and Systems, E-93D, 2, 280-289, 2010.02. |

40. | Kohei Hatano, Eiji Takimoto, Linear Programming Boosting by Column and Row Generation, 12th International Conference on Discovery Science (DS 2009), LNAI, 5808, 401-408, 2009.10. |

41. | Kei Uchizawa, Takao Nishizeki, Eiji Takimoto, Energy Complexity and Depth of Threshold Circuits, 17th International Symposium on Fundamentals of Computation Theory (FCS 2009), LNCS, 5699, 335-345, 2009.09. |

42. | Kei Uchizawa, Takao Nishizeki, Eiji Takimoto, Size and Energy of Threshold Circuits Computing Mod Functions, 34th International Symposium on Mathematical Foundations of Computer Science (MFCS 2009), LNCS, 5734, 724-735, 2009.09. |

43. | Hideaki Fukuhara, Eiji Takimoto, Lower Bounds on Quantum Query Complexity for Read-once Formulas with XOR and MUX Operators , Proceedings of the 2nd Annual Meeting of Asian Association for Algorithms and Computation (AAAC 2009), 8-8, 2009.04. |

44. | Hideaki Fukuhara, Eiji Takimoto, Lower bounds on quantum query complexity for read-once decision trees with parity nodes, Computing: The Australasian Theory Symposium (CATS 2009), CRPIT, 94, 474-487, 2009.01. |

45. | Kei Uchizawa, Eiji Takimoto, Exponential lower bounds on the size of constant-depth threshold circuits with small energy complexity, Theoretical Computer Science, 407, 1-3, 474-487, 2008.11. |

46. | Takayuki Sato, Kazuyuki Amano, Eiji Takimoto, Akira Maruoka, Monotone DNF Formula that has a Minimal or Maximal Number of Satisfying Assignments, The 14th Annual International Computing and Combinatorics Conference (COCOON), LNCS, 5092, 191-203, 2008.06. |

47. | Jun-ichi Moribe, Kohei Hatano, Eiji Takimoto, Masayuki Takeda, Smooth Boosting for Margin-Based Ranking, 19th International Conference on Algorithmic Learning Theory (ALT 2008), LNAI, 5254, 227-239, 2008.10. |

48. | Kei Uchizawa, Eiji Takimoto, Energy Complexity of Threshold Circuits, The First AAAC Annual Meeting , pp. 45-45, 2008.04. |

49. | Masashi Karasaki, Eiji Takimoto, On noise-reduction effect of filters for Boolean functions , The First AAAC Anual Meeting, pp.31-31, 2008.04. |

Unauthorized reprint of the contents of this database is prohibited.