Basic Information

Faculty

• 工学部

Academic Degree

• 修士（情報理工学）(東京大学)
• 博士（科学）(東京大学)

Academic Resume (Graduate Schools)

• 東京大学, 大学院情報理工学系研究科数理情報学専攻修士課程, 修了
• 東京大学, 大学院新領域創成科学研究科複雑理工学専攻博士課程, 修了

Academic Resume (Undergraduate School/Majors)

• 東京大学, 工学部計数工学科, 卒業

Research History

• From Jan. 2021, To Present

  Kyoto University, Department of Systems Science, Graduate School of Informatics, Associate Professor
• From Jan. 2017, To Dec. 2020

  The University of Tokyo, Department of Complexity Science and Engineering, Graduate School of Frontier Sciences, Lecturer
• From Mar. 2013, To Dec. 2016

  The University of Tokyo, Department of Complexity Science and Engineering, Graduate School of Frontier Sciences, Research Associate

ID,URL

• J-Global ID

  202001010142324778

researchmap URL

• https://researchmap.jp/jhonda

Research

Research Topics, Overview of the research

• Research Topics

  Machine learning, bandit problems, reinforcement learning.

• Overview of the research

  We are researching online decision making problems in machine learning such as bandit problems and reinforcement learning. For these topics we are studying construction of good algorithms and their limitations from theoretical and practical viewpoints.

Papers

• Follow-the-Perturbed-Leader with Fréchet-type Tail Distributions: Optimality in Adversarial Bandits and Best-of-Both-Worlds.

  Jongyeong Lee; Junya Honda; Shinji Ito; Min-hwan Oh

  COLT, 2024

  
• Adaptive Learning Rate for Follow-the-Regularized-Leader: Competitive Analysis and Best-of-Both-Worlds.

  Shinji Ito; Taira Tsuchiya; Junya Honda

  COLT, 2024

  
• Stability-penalty-adaptive follow-the-regularized-leader: Sparsity, game-dependency, and best-of-both-worlds.

  Taira Tsuchiya; Shinji Ito; Junya Honda

  NeurIPS, 2023

  
• Optimality of Thompson Sampling with Noninformative Priors for Pareto Bandits.

  Jongyeong Lee; Junya Honda; Chao-Kai Chiang; Masashi Sugiyama

  ICML, 2023

  
• Thompson Exploration with Best Challenger Rule in Best Arm Identification.

  Jongyeong Lee; Junya Honda; Masashi Sugiyama

  ACML, 2023

  
• Further Adaptive Best-of-Both-Worlds Algorithm for Combinatorial Semi-Bandits.

  Taira Tsuchiya; Shinji Ito; Junya Honda

  AISTATS, 2023

  
• Best-of-Both-Worlds Algorithms for Partial Monitoring.

  Taira Tsuchiya; Shinji Ito; Junya Honda

  ALT, 2023, Peer-reviewed

  
• Follow-the-Perturbed-Leader Achieves Best-of-Both-Worlds for Bandit Problems.

  Junya Honda; Shinji Ito; Taira Tsuchiya

  ALT, 2023, Peer-reviewed

  
• Bayesian optimization with partially specified queries.

  Shogo Hayashi; Junya Honda; Hisashi Kashima

  Machine Learning, 2022, Peer-reviewed

  
• Minimax Optimal Algorithms for Fixed-Budget Best Arm Identification.

  Junpei Komiyama; Taira Tsuchiya; Junya Honda

  NeurIPS, 2022, Peer-reviewed

  
• Nearly Optimal Best-of-Both-Worlds Algorithms for Online Learning with Feedback Graphs.

  Shinji Ito; Taira Tsuchiya; Junya Honda

  NeurIPS, 2022, Peer-reviewed

  
• Adversarially Robust Multi-Armed Bandit Algorithm with Variance-Dependent Regret Bounds.

  Shinji Ito; Taira Tsuchiya; Junya Honda

  COLT, 2022, Peer-reviewed

  
• Bandit Algorithms Based on Thompson Sampling for Bounded Reward Distributions.

  Charles Riou; Junya Honda

  Algorithmic Learning Theory(ALT), 2020, Peer-reviewed

  
• Optimal dose escalation methods using deep reinforcement learning in phase I oncology trials.

  Kentaro Matsuura; Kentaro Sakamaki; Junya Honda; Takashi Sozu

  Journal of biopharmaceutical statistics, 30 Jan. 2023, Peer-reviewed

  
• Polynomial-time Algorithms for Combinatorial Pure Exploration with Full-bandit Feedback.

  Yuko Kuroki; Liyuan Xu; Atsushi Miyauchi 0001; Junya Honda; Masashi Sugiyama

  CoRR, 2019

  
• Optimal adaptive allocation using deep reinforcement learning in a dose-response study.

  Kentaro Matsuura; Junya Honda; Imad El Hanafi; Takashi Sozu; Kentaro Sakamaki

  Statistics in medicine, 07 Nov. 2021, Peer-reviewed

  
• Mediated Uncoupled Learning: Learning Functions without Direct Input-output Correspondences.

  Ikko Yamane; Junya Honda; Florian Yger; Masashi Sugiyama

  Proceedings of the 38th International Conference on Machine Learning(ICML), 2021, Peer-reviewed

  
• Analysis and Design of Thompson Sampling for Stochastic Partial Monitoring.

  Taira Tsuchiya; Junya Honda; Masashi Sugiyama

  Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020(NeurIPS), 2020, Peer-reviewed

  
• Online Dense Subgraph Discovery via Blurred-Graph Feedback.

  Yuko Kuroki; Atsushi Miyauchi 0001; Junya Honda; Masashi Sugiyama

  Proceedings of the 37th International Conference on Machine Learning(ICML), 2020, Peer-reviewed

  
• Polynomial-Time Algorithms for Multiple-Arm Identification with Full-Bandit Feedback.

  Yuko Kuroki; Liyuan Xu; Atsushi Miyauchi 0001; Junya Honda; Masashi Sugiyama

  Neural Comput., 2020, Peer-reviewed

  
• A bad arm existence checking problem: How to utilize asymmetric problem structure?

  Tabata, Koji; Nakamura, Atsuyoshi; Honda, Junya; Komatsuzaki, Tamiki

  MACHINE LEARNING, Oct. 2019, Peer-reviewed

  
• Good arm identification via bandit feedback

  Hideaki Kano; Junya Honda; Kentaro Sakamaki; Kentaro Matsuura; Atsuyoshi Nakamura; Masashi Sugiyama

  MACHINE LEARNING, May 2019, Peer-reviewed

  
• Uncoupled Regression from Pairwise Comparison Data.

  Liyuan Xu; Junya Honda; Gang Niu; Masashi Sugiyama

  Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019(NeurIPS), 2019, Peer-reviewed

  
• On the Calibration of Multiclass Classification with Rejection.

  Chenri Ni; Nontawat Charoenphakdee; Junya Honda; Masashi Sugiyama

  Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019(NeurIPS), 2019, Peer-reviewed

  
• Learning from Positive and Unlabeled Data with a Selection Bias.

  Masahiro Kato; Takeshi Teshima; Junya Honda

  7th International Conference on Learning Representations, 2019, Peer-reviewed

  
• Unsupervised Domain Adaptation Based on Source-Guided Discrepancy

  Seiichi Kuroki; Nontawat Charoenphakdee; Han Bao; Junya Honda; Issei Sato; Masashi Sugiyama

  THIRTY-THIRD AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FIRST INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / NINTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2019, Peer-reviewed

  
• Dueling Bandits with Qualitative Feedback

  Liyuan Xu; Junya Honda; Masashi Sugiyama

  THIRTY-THIRD AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FIRST INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / NINTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2019, Peer-reviewed

  
• Exploring a potential energy surface by machine learning for characterizing atomic transport

  Kanamori, Kenta; Toyoura, Kazuaki; Honda, Junya; Hattori, Kazuki; Seko, Atsuto; Karasuyama, Masayuki; Shitara, Kazuki; Shiga, Motoki; Kuwabara, Akihide; Takeuchi, Ichiro

  PHYSICAL REVIEW B, Mar. 2018, Peer-reviewed

  
• Nonconvex Optimization for Regression with Fairness Constraints.

  Junpei Komiyama; Akiko Takeda; Junya Honda; Hajime Shimao

  Proceedings of the 35th International Conference on Machine Learning(ICML), 2018, Peer-reviewed

  
• A fully adaptive algorithm for pure exploration in linear bandits.

  Liyuan Xu; Junya Honda; Masashi Sugiyama

  International Conference on Artificial Intelligence and Statistics(AISTATS), 2018, Peer-reviewed

  
• Normal Bandits of Unknown Means and Variances

  Cowan, Wesley; Honda, Junya; Katehakis, Michael N.

  JOURNAL OF MACHINE LEARNING RESEARCH, 2018, Peer-reviewed

  
• Exact Asymptotics of Random Coding Error Probability for General Memoryless Channels

  Honda, Junya

  2018 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2018, Peer-reviewed

  
• Worst-case Redundancy of Optimal Binary AIFV Codes and Their Extended Codes

  Hu, Weihua; Yamamoto, Hirosuke; Honda, Junya

  IEEE TRANSACTIONS ON INFORMATION THEORY, Aug. 2017, Peer-reviewed

  
• Variable-to-Fixed Length Homophonic Coding Suitable for Asymmetric Channel Coding

  Honda, Junya; Yamamoto, Hirosuke

  2017 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2017, Peer-reviewed

  
• Position-based Multiple-play Bandit Problem with Unknown Position Bias.

  Junpei Komiyama; Junya Honda; Akiko Takeda

  ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 30 (NIPS 2017), 2017, Peer-reviewed

  
• Copeland Dueling Bandit Problem: Regret Lower Bound, Optimal Algorithm, and Computationally Efficient Algorithm.

  Junpei Komiyama; Junya Honda; Hiroshi Nakagawa

  Proceedings of the 33nd International Conference on Machine Learning(ICML), 2016, Peer-reviewed

  
• Variable-to-Fixed Length Homophonic Coding with a Modified Shannon-Fano-Flias Code

  Honda, Junya; Yamamoto, Hirosuke

  PROCEEDINGS OF 2016 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2016), 2016, Peer-reviewed

  
• Tight Upper Bounds on the Redundancy of Optimal Binary AIFV Codes

  Hu, Weihua; Yamamoto, Hirosuke; Honda, Junya

  2016 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, 2016, Peer-reviewed

  
• Non-Asymptotic Analysis of a New Bandit Algorithm for Semi-Bounded Reward

  Honda, Junya; Takemura, Akimichi

  JOURNAL OF MACHINE LEARNING RESEARCH, Dec. 2015, Peer-reviewed

  
• Almost Instantaneous Fixed-to-Variable Length Codes

  Yamamoto, Hirosuke; Tsuchihashi, Masato; Honda, Junya

  IEEE TRANSACTIONS ON INFORMATION THEORY, Dec. 2015, Peer-reviewed

  
• Optimal Regret Analysis of Thompson Sampling in Stochastic Multi-armed Bandit Problem with Multiple Plays.

  Junpei Komiyama; Junya Honda; Hiroshi Nakagawa

  Proceedings of the 32nd International Conference on Machine Learning(ICML), 2015, Peer-reviewed

  
• Regret Lower Bound and Optimal Algorithm in Dueling Bandit Problem.

  Junpei Komiyama; Junya Honda; Hisashi Kashima; Hiroshi Nakagawa

  Proceedings of The 28th Conference on Learning Theory(COLT), 2015, Peer-reviewed

  
• Exact Asymptotics for the Random Coding Error Probability

  Honda, Junya

  2015 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2015, Peer-reviewed

  
• Regret Lower Bound and Optimal Algorithm in Finite Stochastic Partial Monitoring

  Komiyama, Junpei; Honda, Junya; Nakagawa, Hiroshi

  ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 28 (NIPS 2015), 2015, Peer-reviewed

  
• FV Polar Coding for Lossy Compression with an Improved Exponent

  Wang, Runxin; Honda, Junya; Yamamoto, Hirosuke; Liu, Rongke

  2015 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2015, Peer-reviewed

  
• Construction of Polar Codes for Channels with Memory

  Wang, Runxin; Honda, Junya; Yamamoto, Hirosuke; Liu, Rongke; Hou, Yi

  2015 IEEE INFORMATION THEORY WORKSHOP - FALL (ITW), 2015, Peer-reviewed

  
• Variable Length Lossy Coding Using an LDPC Code

  Honda, Junya; Yamamoto, Hirosuke

  IEEE TRANSACTIONS ON INFORMATION THEORY, Jan. 2014, Peer-reviewed

  
• Optimality of Thompson Sampling for Gaussian Bandits Depends on Priors

  Honda, Junya; Takemura, Akimichi

  ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 33, 2014, Peer-reviewed

  
• RSA Meets DPA: Recovering RSA Secret Keys from Noisy Analog Data

  Kunihiro, Noboru; Honda, Junya

  CRYPTOGRAPHIC HARDWARE AND EMBEDDED SYSTEMS - CHES 2014, 2014, Peer-reviewed

  
• Polar Coding Without Alphabet Extension for Asymmetric Models

  Honda, Junya; Yamamoto, Hirosuke

  IEEE TRANSACTIONS ON INFORMATION THEORY, Dec. 2013, Peer-reviewed

  
• Stochastic bandit based on empirical moments

  Junya Honda; Akimichi Takemura

  Journal of Machine Learning Research, 2012, Peer-reviewed

  
• Polar Coding without Alphabet Extension for Asymmetric Channels

  Honda, Junya; Yamamoto, Hirosuke

  2012 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY PROCEEDINGS (ISIT), 2012, Peer-reviewed

  
• Fast Linear-Programming Decoding of LDPC Codes over GF(2(m))

  Honda, Junya; Yamamoto, Hirosuke

  2012 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS (ISITA 2012), 2012, Peer-reviewed

  
• An asymptotically optimal policy for finite support models in the multiarmed bandit problem

  Honda, Junya; Takemura, Akimichi

  MACHINE LEARNING, Dec. 2011, Peer-reviewed

  
• An asymptotically optimal bandit algorithm for bounded support models

  Junya Honda; Akimichi Takemura

  COLT 2010 - The 23rd Conference on Learning Theory, 2010, Peer-reviewed

  
• Variable Length Lossy Coding using an LDPC Code

  Honda, Junya; Yamamoto, Hirosuke

  2009 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, VOLS 1- 4, 2009, Peer-reviewed

  
• Application of LCLP to Lossy Source Coding

  Miyake, Shigeki; Honda, Junya; Yamamoto, Hirosuke

  2008 INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY AND ITS APPLICATIONS, VOLS 1-3, 2008, Peer-reviewed

  

Misc.

• Construction of Almost Instantaneous FV codes based on Integer Programming

  TSUCHIHASHI Masato; YAMAMOTO Hirosuke; HONDA Junya

  IEICE technical report. Information theory, 17 Jul. 2014

  
• Variable Length Coding of Asymmetric Channels using a Homophonic Code

  HONDA Junya; YAMAMOTO Hirosuke

  IEICE technical report. Information theory, 12 Jul. 2012

  
• On a Non-asymptotic Analysis Using Large Deviation Principles in the Multiarmed Bandit Problem

  HONDA Junya; TAKEMURA Akimichi

  電子情報通信学会技術研究報告. IBISML, 情報論的学習理論と機械学習 = IEICE technical report. IBISML, Information-based induction sciences and machine learning, 12 Jun. 2012

  
• Bad Arm Existence Checking Algorithms with Small Sample Complexity

  田畑 公次; 中村 篤祥; 本多 淳也; 小松崎 民樹

  人工知能基本問題研究会, 16 Mar. 2018

  
• Source Coding with LDPC Code using Linear Programming

  HONDA Junya; MIYAKE Shigeki; YAMAMOTO Hirosuke; MARUYAMA Mitsuru

  IEICE technical report, 17 Jul. 2008

  

Books and Other Publications

• バンディット問題の理論とアルゴリズム = Theory and algorithms for bandit problems

  本多, 淳也; 中村, 篤祥; 講談社サイエンティフィク

  講談社, 2016

  Find Kyoto University's collection of books :

  

External funds: Kakenhi

• バンディット問題における最適性達成のためのランダム方策の発展と解析

  Grant-in-Aid for Scientific Research (C)

  Basic Section 60010:Theory of informatics-related

  Kyoto University

  本多 淳也

  From 01 Apr. 2021, To 31 Mar. 2024, Granted

  機械学習;学習理論;実験計画;情報理論

  https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-21K11747/
• Construction of knowledge discovery algorithms based on information theoretic methods

  Grant-in-Aid for Early-Career Scientists

  Basic Section 60010:Theory of informatics-related

  Kyoto University;The University of Tokyo

  Junya Honda

  From 01 Apr. 2018, To 31 Mar. 2021, Project Closed

  機械学習;情報理論;バンディット問題

  https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-18K17998/
• 確率モデルに基づいた結晶構造の学習および推定法の確立

  Grant-in-Aid for Scientific Research on Innovative Areas (Research in a proposed research area)

  Science and Engineering

  The University of Tokyo

  本多 淳也

  From 01 Apr. 2016, To 31 Mar. 2018, Project Closed

  機械学習;実験計画法;材料探索;材料科学;計算物理

  https://kaken.nii.ac.jp/grant/KAKENHI-PUBLICLY-16H00881/
• 効率的な材料探索のための実験計画法の開発

  Grant-in-Aid for Scientific Research on Innovative Areas (Research in a proposed research area)

  Science and Engineering

  The University of Tokyo

  本多 淳也

  From 01 Apr. 2014, To 31 Mar. 2016, Project Closed

  機械学習

  https://kaken.nii.ac.jp/grant/KAKENHI-PUBLICLY-26106506/
• 多腕バンディット問題における最適戦略の構成と発展

  Grant-in-Aid for Research Activity Start-up

  The University of Tokyo

  本多 淳也

  From 30 Aug. 2013, To 31 Mar. 2015, Declined

  機械学習;統計学;多腕バンディット問題

  https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-25880006/

Education

Teaching subject(s)

• From 01 Apr. 2024, To 31 Mar. 2025

  Mathematical theory of information and communications

  9142, Fall, Faculty of Engineering, 2
• From 01 Apr. 2024, To 31 Mar. 2025

  Introduction to Applied Mathematics and Physics

  9114, Spring, Faculty of Engineering, 2
• From 01 Apr. 2024, To 31 Mar. 2025

  Stochastic Discrete Event Systems

  9096, Spring, Faculty of Engineering, 2
• From 01 Apr. 2024, To 31 Mar. 2025

  Systems Science, Advanced II

  3512, Fall, Graduate School of Informatics, 2
• From 01 Apr. 2024, To 31 Mar. 2025

  Systems Science, Advanced I

  3542, Spring, Graduate School of Informatics, 2
• From 01 Apr. 2024, To 31 Mar. 2025

  Statistical Systems Theory

  3536, Spring, Graduate School of Informatics, 2
• From 01 Apr. 2024, To 31 Mar. 2025

  Statistical Systems Theory

  M310, Spring, Institute for Liberal Arts and Sciences, 2
• From 01 Apr. 2023, To 31 Mar. 2024

  Stochastic Discrete Event Systems

  9096, Spring, Faculty of Engineering, 2
• From 01 Apr. 2023, To 31 Mar. 2024

  Mathematical theory of information and communications

  9142, Fall, Faculty of Engineering, 2
• From 01 Apr. 2023, To 31 Mar. 2024

  Systems Science, Advanced I

  3542, Spring, Graduate School of Informatics, 2
• From 01 Apr. 2023, To 31 Mar. 2024

  Statistical Systems Theory

  3536, Spring, Graduate School of Informatics, 2
• From 01 Apr. 2023, To 31 Mar. 2024

  Systems Science, Advanced II

  3512, Fall, Graduate School of Informatics, 2
• From 01 Apr. 2023, To 31 Mar. 2024

  ILAS Seminar

  Z001, Spring, Institute for Liberal Arts and Sciences, 2
• From 01 Apr. 2023, To 31 Mar. 2024

  Statistical Systems Theory

  M310, Spring, Institute for Liberal Arts and Sciences, 2
• From 01 Apr. 2022, To 31 Mar. 2023

  Mathematical theory of information and communications

  9142, Fall, Faculty of Engineering, 2
• From 01 Apr. 2022, To 31 Mar. 2023

  Seminar on Applied Mathematics and Physics

  9074, Fall, Faculty of Engineering, 2
• From 01 Apr. 2022, To 31 Mar. 2023

  Systems Science, Advanced I

  3542, Spring, Graduate School of Informatics, 2
• From 01 Apr. 2022, To 31 Mar. 2023

  Systems Science, Advanced II

  3512, Fall, Graduate School of Informatics, 2
• From 01 Apr. 2022, To 31 Mar. 2023

  Statistical Systems Theory

  3536, Spring, Graduate School of Informatics, 2
• From 01 Apr. 2022, To 31 Mar. 2023

  Statistical Systems Theory

  M310, Spring, Institute for Liberal Arts and Sciences, 2
• From 01 Apr. 2022, To 31 Mar. 2023

  Stochastic Discrete Event Systems

  9096, Spring, Faculty of Engineering, 2
• From Apr. 2021, To Mar. 2022

  Mathematical theory of information and communications

  Fall, 工学部
• From Apr. 2021, To Mar. 2022

  Stochastic Discrete Event Systems

  Spring, 工学部
• From Apr. 2021, To Mar. 2022

  Systems Science, Advanced II

  Fall, 情報学研究科
• From Apr. 2021, To Mar. 2022

  Systems Science, Advanced I

  Spring, 情報学研究科
• From Apr. 2021, To Mar. 2022

  Statistical Systems Theory

  Spring, 情報学研究科

Participation in PhD Defense

• Environment Adaptive Regret Analysis in Bandit Problems

  TSUCHIYA TAIRA, Graduate School of Informatics, Chief Examiner

  25 Sep. 2023
• Convergence Formulas for the Level-increment Truncation Approximation of M/G/1-type Markov Chains

  OUCHI KATSUHISA, Graduate School of Informatics, Sub-chief Examiner

  24 Nov. 2023

Administration

Faculty management (title, position)

• From 01 Apr. 2021, To 31 Mar. 2023

  広報WG委員