Junya Honda

Associate Professor Department of Systems Science, Graduate School of Informatics, Kyoto University. Center for Advanced Integrated Intelligence Research, RIKEN, Tokyo, Japan. Room 104, Faculty of Engineering Integrated Research Bldg., Kyoto University, Yoshida Honmachi, Sakyo-ku, Kyoto, 606-8501, Japan. TEL: +81-75-753-5509 E-Mail: honda [atmark] i.kyoto-u.ac.jp Research Interests Machine Learning Bandit problems Reinforcement learning Bayesian optimization Information Theory Channel coding Rate-distortion theory Polar codes LDPC codes

Publications

Google Scholar

• Jongyeong Lee, Junya Honda, Shinji Ito, Min-hwan Oh, "Follow-the-Perturbed-Leader with Fréchet-type Tail Distributions: Optimality in Adversarial Bandits and Best-of-Both-Worlds",
  The 37th Annual Conference on Learning Theory (COLT 2024), pp. 3375-3430, June 30 - July 3, 2024.
• Shinji Ito, Taira Tsuchiya, Junya Honda, "Adaptive Learning Rate for Follow-the-Regularized-Leader: Competitive Ratio Analysis and Best-of-Both-Worlds",
  The 37th Annual Conference on Learning Theory (COLT 2024), pp. 2522-2563, June 30 - July 3, 2024.
• Taira Tsuchiya, Shinji Ito, Junya Honda, "Exploration by Optimization with Hybrid Regularizers: Logarithmic Regret with Adversarial Robustness in Partial Monitoring",
  The 41st International Conference on Machine Learning (ICML 2024), pp. 48768-48790, July 21-27, 2024.
• Kazuma Shimizu, Junya Honda, Shinji Ito, Shinji Nakadai, "Learning with Posterior Sampling for Revenue Management under Time-varying Demand",
  The 33rd International Joint Conference on Artificial Intelligence (IJCAI 2024), pp.4911-491, Jeju, Korea, Aug. 3-9, 2024. (full version)
• Taira Tsuchiya, Shinji Ito, Junya Honda, "Stability-penalty-adaptive Follow-the-regularized-leader: Sparsity, Game-dependency, and Best-of-both-worlds",
  The 37th Neural Information Processing Systems (NeurIPS2023), pp.47406-47437, New Orleans, USA, Dec. 10-16, 2023.
• Jongyeong Lee, Junya Honda, Masashi Sugiyama, "Thompson Exploration with Best Challenger Rule in Best Arm Identification",
  The 15th Asian Conference on Machine Learning (ACML 2023), pp. 646-661, Istanbul, Turkey, Nov. 11-14.
• Jongyeong Lee, Junya Honda, Chao-Kai Chiang, Masashi Sugiyama, "Optimality of Thompson Sampling with Noninformative Priors for Pareto Bandits",
  The 40th International Conference on Machine Learning (ICML2023), pp.18810-18851, Hawaii, USA, July 23-29, 2023.
• Taira Tsuchiya, Shinji Ito, Junya Honda, "Further Adaptive Best-of-Both-Worlds Algorithm for Combinatorial Semi-Bandits",
  The 26th International Conference on Artificial Intelligence and Statistics (AISTATS2023), pp. 8117-8144, Valencia, Spain, Apr. 25-27, 2023.
• Junya Honda, Shinji Ito, Taira Tsuchiya, "Follow-the-Perturbed-Leader Achieves Best-of-Both-Worlds for Bandit Problems",
  The 34th International Conference on Algorithmic Learning Theory (ALT2023), pp.726-754, Singapore, Feb. 20-23, 2023. (slide)
• Taira Tsuchiya, Shinji Ito, Junya Honda, "Best-of-Both-Worlds Algorithms for Partial Monitoring",
  The 34th International Conference on Algorithmic Learning Theory (ALT2023), pp.1484-1515, Singapore, Feb. 20-23, 2023.
• Shinji Ito, Taira Tsuchiya, Junya Honda, "Nearly Optimal Best-of-Both-Worlds Algorithms for Online Learning with Feedback Graphs",
  The 36th Neural Information Processing Systems (NeurIPS2022), pp.28631-28643, Dec. 28 - Nov. 4, 2022.
• Junpei Komiyama, Taira Tsuchiya, Junya Honda, "Minimax Optimal Algorithms for Fixed-Budget Best Arm Identification",
  The 36th Neural Information Processing Systems (NeurIPS2022), pp.10393-10404, Dec. 28 - Nov. 4, 2022.
• Shinji Ito, Taira Tsuchiya, Junya Honda, "Adversarially Robust Multi-Armed Bandit Algorithm with Variance-Dependent Regret Bounds",
  The 35th Annual Conference on Learning Theory (COLT2022), extended abstract. (full version)
• Ikko Yamane, Junya Honda, Florian Yger, Masashi Sugiyama, "Mediated Uncoupled Learning: Learning Functions without Direct Input-output Correspondences",
  The 38th International Conference on Machine Learning (ICML2021), pp.11637-11647, July 18-24, 2021.
• Taira Tsuchiya, Junya Honda, Masashi Sugiyama, "Analysis and Design of Thompson Sampling for Stochastic Partial Monitoring",
  The 34th Neural Information Processing Systems (NeurIPS2020), pp.8861-8871, Dec. 6-12, 2020.
• Yuko Kuroki, Atsushi Miyauchi, Junya Honda, Masashi Sugiyama, "Online Dense Subgraph Discovery via Blurred-Graph Feedback",
  pp.5522-5532, The 37th International Conference on Machine Learning (ICML2020), July 13-18, 2020.
• Charles Riou, Junya Honda, "Bandit Algorithms Based on Thompson Sampling for Bounded Reward Distributions",
  The 31st International Conference on Algorithmic Learning Theory (ALT2020), pp.777-826, San Diego, USA, Feb. 8-11, 2020.
• Liyuan Xu, Junya Honda, Gang Niu, Masashi Sugiyama, "Uncoupled Regression from Pairwise Comparison Data",
  The 33rd Neural Information Processing Systems (NeurIPS2019), pp.3992-4002, Vancouver, Canada, Dec. 8-14, 2019.
• Chenri Ni, Nontawat Charoenphakdee, Junya Honda, Masashi Sugiyama, "On the Calibration of Multiclass Classification with Rejection",
  The 33rd Neural Information Processing Systems (NeurIPS2019), pp.2586-2596, Vancouver, Canada, Dec. 8-14, 2019.
• Masahiro Kato, Teshima Takeshi, Junya Honda, "Learning from Positive and Unlabeled Data with a Selection Bias",
  The Seventh International Conference on Learning Representations (ICLR2019), New Orleans, USA, May. 6-9, 2019.
• Seiichi Kuroki, Nontawat Charoenphakdee, Han Bao, Junya Honda, Issei Sato, Masashi Sugiyama, "Unsupervised Domain Adaptation Based on Source-guided Discrepancy",
  The 23rd AAAI Conference on Artificial Intelligence (AAAI2019), pp.5549-5556, Honolulu, USA, Jan. 27-Feb. 1, 2019.
• Liyuan Xu, Junya Honda, Masashi Sugiyama, "Dueling Bandits with Qualitative Feedback",
  The 23rd AAAI Conference on Artificial Intelligence (AAAI2019), pp.4122-4129, Honolulu, USA, Jan. 27-Feb. 1, 2019.
• Junya Honda, "Exact Asymptotics of Random Coding Error Probability for General Memoryless Channels",
  2018 IEEE International Symposium on Information Theory (ISIT2018), pp.1844-1848, Vail, USA, June 17-22, 2018.
  TPC choice paper (7% of accepted papers). (full version)
• Junpei Komiyama, Akiko Takeda, Junya Honda, Hajime Shimao, "Nonconvex Optimization for Regression with Fairness Constraints",
  The 35th International Conference on Machine Learning (ICML2018), pp.2737-2746, Stockholm, Sweden, July 10-15, 2018.
• Liyuan Xu, Junya Honda, Masashi Sugiyama, "A fully adaptive algorithm for pure exploration in linear bandits",
  The 21st International Conference on Artificial Intelligence and Statistics (AISTATS2018), pp. 843-851, Canary Islands, Spain, Apr. 9-11, 2018.
• Junpei Komiyama, Junya Honda, Akiko Takeda, "Position-based Multiple-play Multi-armed Bandit Problem with Unknown Position Bias",
  The 31st Neural Information Processing Systems (NIPS2017), pp.5005-5015, Long Beach, USA, Dec. 4-9, 2017.
• Junya Honda, Hirosuke Yamamoto, "Variable-to-fixed length homophonic coding suitable for asymmetric channel coding",
  2017 IEEE International Symposium on Information Theory (ISIT2017), pp.2860-2864, Aachen, Germany, June 25-30, 2017.
• Junya Honda, Hirosuke Yamamoto, "Variable-to-Fixed Length Homophonic Coding with a Modified Shannon-Fano-Elias Code".
  2016 International Symposium on Information Theory and its Applications (ISITA2016), pp.11-15, Monterey, USA, Oct.10-Nov.2, 2016.
• Weihua Hu, Hirosuke Yamamoto, Junya Honda, "Tight Upper Bounds on the Redundancy of Optimal Binary AIFV Codes",
  2016 IEEE International Symposium on Information Theory (ISIT2016), pp.6-10, Barcelona, Spain, July 10-15, 2016.
• Junpei Komiyama, Junya Honda, Hiroshi Nakagawa, "Copeland Dueling Bandit Problem: Regret Lower Bound, Optimal Algorithm, and Computationally Efficient Algorithm".
  The 33rd International Conference on Machine Learning (ICML2016), pp.1235-1244, New York, USA, June 19-24, 2016.
• Junpei Komiyama, Junya Honda, Hiroshi Nakagawa, "Regret Lower Bound and Optimal Algorithm in Finite Stochastic Partial Monitoring".
  The 29th Neural Information Processing Systems (NIPS2015), pp.1792-1800, Montreal, Canada, Dec. 7-12, 2015.
• Runxin Wang, Junya Honda, Hirosuke Yamamoto, Rongke Liu, Yi Hou, "Construction of Polar Codes for Channels with Memory",
  IEEE Information Theory Workshop (ITW2015), pp.187-191, Jeju Island, Korea, Oct. 11-15, 2015.
• Junpei Komiyama, Junya Honda, Hiroshi Nakagawa, "Optimal Regret Analysis of Thompson Sampling in Stochastic Multi-armed Bandit Problem with Multiple Plays".
  The 32nd International Conference on Machine Learning (ICML2015), pp.1152-1161, Lille, France, July 6-11, 2015.
• Junpei Komiyama, Junya Honda, Hisashi Kashima, Hiroshi Nakagawa, "Regret Lower Bound and Optimal Algorithm in Dueling Bandit Problem".
  The 28th Annual Conference on Learning Theory (COLT2015), pp.1141-1154, Paris, France, July 3-6, 2015.
• Junya Honda, "Exact Asymptotics for Random Coding Error Probability",
  2015 IEEE International Symposium on Information Theory (ISIT2015), pp.91-95, Hong Kong, June 14-19, 2015.
• Runxin Wang, Junya Honda, Hirosuke Yamamoto, "FV Polar Coding for Lossy Compression with an Improved Exponent",
  2015 IEEE International Symposium on Information Theory (ISIT2015), pp.1517-1521, Hong Kong, June 14-19, 2015.
• Noboru Kunihiro, Junya Honda, "RSA meets DPA: Recovering RSA Secret Keys from Noisy Analog Data",
  Workshop on Cryptographic Hardware and Embedded Systems 2014 (CHES 2014), pp.261-278, Busan, Korea, Sep. 23-26, 2014.
• Junya Honda, Akimichi Takemura, "Optimality of Thompson Sampling for Gaussian Bandits Depends on Priors".
  Seventeenth International Conference on Artificial Intelligence and Statistics (AISTATS2014), pp.375-383, Reykjavik, Iceland, Apr. 22-25, 2014.
• Junya Honda, Hirosuke Yamamoto, "Fast Linear-Programming Decoding of LDPC Codes over GF(2^m)",
  2012 International Symposium on Information Theory and its Applications (ISITA 2012), pp.754-758, Honolulu, USA, Oct. 28-31, 2012.
• Junya Honda, Hirosuke Yamamoto, "Polar Coding without Alphabet Extension for Asymmetric Channels",
  2012 IEEE International Symposium on Information Theory (ISIT2012), pp.2157-2161, Cambridge, USA, July 1-6, 2012.
• Junya Honda, Akimichi Takemura, "Stochastic Bandit Based on Empirical Moments".
  Fifteenth International Conference on Artificial Intelligence and Statistics (AISTATS2012), pp.529-537, Canary Islands, Spain, Apr. 21-23, 2012.
• Junya Honda, Akimichi Takemura, "An Asymptotically Optimal Bandit Algorithm for Bounded Support Models".
  The 23rd Annual Conference on Learning Theory (COLT2010), pp.67-79, Haifa, Israel, June 27-29, 2010.
• Junya Honda, Hirosuke Yamamoto, "Variable Length Lossy Coding using an LDPC Code",
  2009 IEEE International Symposium on Information Theory (ISIT2009), pp.1973-1977, Seoul, Korea, June 28-July 3, 2009.
• Shigeki Miyake, Junya Honda, Hirosuke Yamamoto, "Application of LCLP to Lossy Source Coding",
  2008 International Symposium on Information Theory and its Applications (ISITA2008), pp.589-594, Auckland, New Zealand, Dec. 2008.

• Charles Riou, Junya Honda, Masashi Sugiyama, "The Survival Bandit Problem", Transactions on Machine Learning Research, 2024.
• Junpei Komiyama, Edouard Fouché, Junya Honda, "Finite-time Analysis of Globally Nonstationary Multi-Armed Bandits",
  Journal of Machine Learning Research, vol.25(112), pp. 1-56, 2024.
• Kentaro Matsuura, Kentaro Sakamaki, Junya Honda, Takashi Sozu, "Optimal dose escalation methods using deep reinforcement learning in phase I oncology trials",
  Journal of Biopharmaceutical Statistics, vol.33, no.5, pp.639-652, 2023.
• Mugihiko Kato, Toshiki Okumura, Yasuhiro Tsubo, Junya Honda, Masashi Sugiyama, Kazushige Touhara, Masako Okamoto, "Spatiotemporal dynamics of odor representations in the human brain revealed by EEG decoding",
  Proceedings of the National Academy of Sciences (PNAS), vol.119, no.21, e2114966119, May 2022.
• Shogo Hayashi, Junya Honda, Hisashi Kashima, "Bayesian optimization with partially specified queries",
  Machine Learning, vol.111, no.3, 1019-1048, Jan. 2022.
• Kentaro Matsuura, Junya Honda, Imad El Hanafi, Takashi Sozu, Kentaro Sakamaki, "Optimal adaptive allocation using deep reinforcement learning in a dose-response study"
  Statistics in Medicine, vol.41, no.7, pp.1157-1171, March. 2022.
• Kohei Shinohara, Atsuto Seko, Takashi Horiyama, Masakazu Ishihata, Junya Honda and Isao Tanaka, "Enumeration of nonequivalent substitutional structures using advanced data structure of binary decision diagram"
  The Journal of Chemical Physics, vol.153, no.10 pp.104-109, Sep. 2020.
• Yuko Kuroki, Liyuan Xu, Atsushi Miyauchi, Junya Honda, Masashi Sugiyama, "Polynomial-time Algorithms for Multiple-arm Identification with Full-bandit Feedback"
  Neural Computation, vol.32, no.9, pp.1733-1773, Sep. 2020.
• Koji Tabata, Atsuyoshi Nakamura, Junya Honda, Tamiki Komatsuzaki, "A bad arm existence checking problem: How to utilize asymmetric problem structure?"
  Machine Learning, vol.109, no. 2. pp.327-372, Feb. 2019.
• Hideaki Kano, Junya Honda, Kentaro Sakamaki, Kentaro Matsuura, Atsuyoshi Nakamura, Masashi Sugiyama, "Good Arm Identification via Bandit Feedback",
  Machine Learning, vol.105, no.5, pp.721-745, Mar. 2019.
• Wesley Cowan, Junya Honda Michael N. Katehakis, "Normal Bandits of Unknown Means and Variances".
  Journal of Machine Learning Research, vol.18(1), pp.5638-5665, 2017. (accepted in Oct. 2015)
• Kenta Kanamori, Kazuaki Toyoura, Junya Honda, Kazuki Hattori, Atsuto Seko, Masayuki Karasuyama, Kazuki Shitara, Motoki Shiga, Akihide Kuwabara, Ichiro Takeuchi, "Exploring a potential energy surface by machine learning for characterizing atomic transport".
  Physical Review B, vol.97, no.12 (125124), Mar. 2018.
• Weihua Hu, Hirosuke Yamamoto, Junya Honda, "Worst-case Redundancy of Optimal Binary AIFV Codes and Their Extended Codes".
  IEEE Transactions on Information Theory, vol.63, no.8, pp.5074-5086, Aug. 2017.
• Junya Honda, Akimichi Takemura, "Non-Asymptotic Analysis of a New Bandit Algorithm for Semi-Bounded Rewards".
  Journal of Machine Learning Research, vol.16, pp.3721-3756, Dec. 2015.
• Hirosuke Yamamoto, Masato Tsuchihashi, Junya Honda, "Almost Instantaneous Fix-to-Variable Length Codes".
  IEEE Transactions on Information Theory, vol.61, no.12, pp.6432-6443, Dec. 2015.
• Junya Honda, Hirosuke Yamamoto, "Variable Length Lossy Coding Using an LDPC Code".
  IEEE Transactions on Information Theory, vol.60, no.1, pp.762-775, Jan. 2014.
• Junya Honda, Hirosuke Yamamoto, "Polar Coding without Alphabet Extension for Asymmetric Models".
  IEEE Transactions on Information Theory, vol.59, no.12, pp.7829-7838, Dec. 2013.
• Junya Honda, Akimichi Takemura, "An Asymptotically Optimal Policy for Finite Support Models in the Multiarmed Bandit Problem".
  Machine Learning, vol.85, pp.361-391, Dec. 2011.

• "Efficient Polar and LDPC Coding for Asymmetric Channels and Sources (非対称な通信路と情報源の効率的なpolar符号化およびLDPC符号化)", The University of Tokyo, 2013. (pdf)

Education:

• Bachelor of Engineering from Mathematical Engineering Course, Department of Mathematical Engineering and Infomation Physics, Faculty of Engineering, The University of Tokyo, March 2008.
• Master of Information Science and Technology from Department of Mathematical Informatics, Graduate School of Information Science and Technology, The University of Tokyo, March 2010. (supervisor: Professor Akimichi Takemura)
• Ph.D of Science from Department of Complexity Science and Engineering, Graduate School of Frontier Sciences, The University of Tokyo, March 2013. (supervisor: Professor Hirosuke Yamamoto)