Kobayashi Taisuke 小林泰介

Researcher Number	10796452
Other IDs	https://orcid.org/0000-0002-3760-249X
Affiliation (Current)	2024: 国立情報学研究所, 情報学プリンシプル研究系, 助教
Affiliation (based on the past Project Information) *help	2022 – 2024: 国立情報学研究所, 情報学プリンシプル研究系, 助教 2021 – 2022: 奈良先端科学技術大学院大学, 先端科学技術研究科, 客員助教 2018 – 2021: 奈良先端科学技術大学院大学, 先端科学技術研究科, 助教 2017: 奈良先端科学技術大学院大学, 情報科学研究科, 助教
Review Section/Research Field	Principal Investigator Medium-sized Section 61:Human informatics and related fields / Transformative Research Areas, Section (IV) / Basic Section 61050:Intelligent robotics-related / Intelligent robotics Except Principal Investigator Basic Section 21040:Control and system engineering-related
Keywords	Principal Investigator 強化学習 / 最適性 / 内部モデル / ベイズ理論 / モデル同定 / アンサンブルモデル / TD誤差 / 世界モデル / ヒューマンロボットインタラクション / 確率的勾配降下法 … More / 潜在空間抽出 / 深層学習 / 知能ロボティックス / 歩行 / 継続学習 / 多目的最適化 / 知能ロボティクス … More Except Principal Investigator 事象駆動 / サンプル値制御 / 線形行列不等式 / サンプル値制御系 / 切り替え系 / 適応学習 / ネットワーク制御 / 2自由度制御 / 適応学習制御 / ハイブリッドシステム / 切替系 / ネットワーク / 安定性 / 切替えシステム Less

Elucidating Functional Roles of Nonlinear TD Errors Guided by Reinforcement Learning as InferencePrincipal Investigator
- Principal Investigator
  
  小林泰介
- Project Period (FY)
  2024 – 2025
- Research Category
  
  Grant-in-Aid for Transformative Research Areas (A)
- Review Section
  
  Transformative Research Areas, Section (IV)
- Research Institution
  National Institute of Informatics
Efficient and rational prediction process integrated with optimality-conditioned internal modelPrincipal Investigator
- Principal Investigator
  
  小林泰介
- Project Period (FY)
  2024 – 2025
- Research Category
  
  Grant-in-Aid for Challenging Research (Exploratory)
- Review Section
  
  Medium-sized Section 61:Human informatics and related fields
- Research Institution
  National Institute of Informatics
計測・通信品質が保証されない環境下の多目的フィードフォワード最適制御と強化学習
- Principal Investigator
  
  杉本謙二
- Project Period (FY)
  2021 – 2024
- Research Category
  
  Grant-in-Aid for Scientific Research (B)
- Review Section
  
  Basic Section 21040:Control and system engineering-related
- Research Institution
  Nanzan University
  Nara Institute of Science and Technology
Development of deep learning to reveal physical human-robot interaction and its application to safe robot controlPrincipal Investigator
- Principal Investigator
  
  Kobayashi Taisuke
- Project Period (FY)
  2020 – 2023
- Research Category
  
  Grant-in-Aid for Scientific Research (B)
- Review Section
  
  Basic Section 61050:Intelligent robotics-related
- Research Institution
  National Institute of Informatics
  Nara Institute of Science and Technology
Event-trigger-type two-degree-of-freedom control system without quality assurance of measurement/communication
- Principal Investigator
  
  Sugimoto Kenji
- Project Period (FY)
  2018 – 2021
- Research Category
  
  Grant-in-Aid for Scientific Research (B)
- Review Section
  
  Basic Section 21040:Control and system engineering-related
- Research Institution
  Nara Institute of Science and Technology
Locomotion Control for Legged Robot using Hierarchical Multi-Objective Reinforcement LearningPrincipal Investigator
- Principal Investigator
  
  Kobayashi Taisuke
- Project Period (FY)
  2017 – 2019
- Research Category
  
  Grant-in-Aid for Young Scientists (B)
- Research Field
  
  Intelligent robotics
- Research Institution
  Nara Institute of Science and Technology

All 2024 2023 2022 2021 2020 2019 2018 2017

All Journal Article Presentation

[Journal Article] Constrained footstep planning using model-based reinforcement learning in virtual constraint-based walking2024
- Author(s)
  Jin Takanori、Kobayashi Taisuke、Matsubara Takamitsu
- Journal Title
  
  Advanced Robotics
  
  Volume: 38 Issue: 8 Pages: 525-545
- DOI
  10.1080/01691864.2024.2336253
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] Sparse representation learning with modified q-VAE towards minimal realization of world model2023
- Author(s)
  Kobayashi Taisuke、Watanuki Ryoma
- Journal Title
  
  Advanced Robotics
  
  Volume: 37 Issue: 13 Pages: 807-827
- DOI
  10.1080/01691864.2023.2221715
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] AdaTerm: Adaptive T-distribution estimated robust moments for Noise-Robust stochastic gradient optimization2023
- Author(s)
  Ilboudo Wendyam Eric Lionel、Kobayashi Taisuke、Matsubara Takamitsu
- Journal Title
  
  Neurocomputing
  
  Volume: 557 Pages: 126692-126692
- DOI
  10.1016/j.neucom.2023.126692
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] Reward bonuses with gain scheduling inspired by iterative deepening search2023
- Author(s)
  Kobayashi Taisuke
- Journal Title
  
  Results in Control and Optimization
  
  Volume: 12 Pages: 100244-100244
- DOI
  10.1016/j.rico.2023.100244
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] Proximal policy optimization with adaptive threshold for symmetric relative density ratio2023
- Author(s)
  Kobayashi Taisuke
- Journal Title
  
  Results in Control and Optimization
  
  Volume: 10 Pages: 100192-100192
- DOI
  10.1016/j.rico.2022.100192
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] Longitudinal Control for Platooning of Vehicles by Two-degree-of-freedom Structure with Online Tuning2022
- Author(s)
  中尾安宏、小林泰介、杉本謙二
- Journal Title
  
  Transactions of the Society of Instrument and Control Engineers
  
  Volume: 58 Issue: 10 Pages: 443-450
- DOI
  10.9746/sicetr.58.443
- ISSN
  0453-4654, 1883-8189
- Language
  Japanese
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K20948
[Journal Article] オンライン調整を伴う2自由度構成による隊列走行車両の縦方向制御2022
- Author(s)
  中尾安宏、小林泰介、杉本謙二
- Journal Title
  
  計測自動制御学会論文集（掲載決定）
  
  Volume: 58
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-23K20948
[Journal Article] Adaptive and multiple time-scale eligibility traces for online deep reinforcement learning2022
- Author(s)
  Kobayashi Taisuke
- Journal Title
  
  Robotics and Autonomous Systems
  
  Volume: 151 Pages: 104019-104019
- DOI
  10.1016/j.robot.2021.104019
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] Sparse Latent Space Acquisition with Variational Autoencoders Based on Tsallis Statistics2022
- Author(s)
  Watanuki Ryoma、Kobayashi Taisuke、Sugimoto Kenji
- Journal Title
  
  Journal of the Robotics Society of Japan
  
  Volume: 40 Issue: 3 Pages: 251-254
- DOI
  10.7210/jrsj.40.251
- ISSN
  0289-1824, 1884-7145
- Language
  Japanese
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] Latent Representation in Human-Robot Interaction With Explicit Consideration of Periodic Dynamics2022
- Author(s)
  Kobayashi Taisuke、Murata Shingo、Inamura Tetsunari
- Journal Title
  
  IEEE Transactions on Human-Machine Systems
  
  Volume: 52 Issue: 5 Pages: 928-940
- DOI
  10.1109/thms.2022.3182909
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] 拡大Tchebyshev関数を用いた多目的最適化としての潜在ダイナミクスモデルの学習2021
- Author(s)
  武田敏季，小林泰介，杉本謙二
- Journal Title
  
  日本ロボット学会誌
  
  Volume: －
- NAID
  130008117307
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] t-soft update of target network for deep reinforcement learning2021
- Author(s)
  Kobayashi Taisuke, Ilboudo Wendyam Eric Lionel
- Journal Title
  
  Neural Networks
  
  Volume: 136 Pages: 63-71
- DOI
  10.1016/j.neunet.2020.12.023
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] q-VAE for Disentangled Representation Learning and Latent Dynamical Systems2020
- Author(s)
  Kobayashi Taisuke
- Journal Title
  
  IEEE Robotics and Automation Letters
  
  Volume: 5 Issue: 4 Pages: 5669-5676
- DOI
  10.1109/lra.2020.3010206
- NAID
  120006949517
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-20H04265
[Journal Article] Student-t policy in reinforcement learning to acquire global optimum of robot control2019
- Author(s)
  Taisuke Kobayashi
- Journal Title
  
  Applied Intelligence
  
  Volume: 49 Issue: 12 Pages: 4335-4347
- DOI
  10.1007/s10489-019-01510-8
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-17K12759
[Presentation] Consolidated Adaptive T-soft Update for Deep Reinforcement Learning2024
- Author(s)
  Kobayashi Taisuke
- Organizer
  IEEE World Congress on Computational Intelligence
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] Mirror-Descent Inverse Kinematics with Box-constrained Joint Space2023
- Author(s)
  Kobayashi Taisuke、Jin Takanori
- Organizer
  World Congress of the International Federation of Automatic Control
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] Domains as Objectives: Multi-Domain Reinforcement Learning with Convex-Coverage Set Learning for Domain Uncertainty Awareness2023
- Author(s)
  Ilboudo Wendyam Eric Lionel、Kobayashi Taisuke、Matsubara Takamitsu
- Organizer
  IEEE/RSJ International Conference on Intelligent Robots and Systems
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] リカレント分散強化学習によるヒステリシスと個体差に頑健な空気圧人工筋の制御2023
- Author(s)
  岡田颯太，小林泰介，松原崇充
- Organizer
  自律分散システム・シンポジウム
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] L2C2: Locally Lipschitz Continuous Constraint towards Stable and Smooth Reinforcement Learning2022
- Author(s)
  Kobayashi Taisuke
- Organizer
  IEEE/RSJ International Conference on Intelligent Robots and Systems
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] リミットサイクル型歩行における長期予測精度の検証2022
- Author(s)
  神孝典，小林泰介，松原崇充
- Organizer
  日本ロボット学会学術講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] Reformulating Multi-Domain Reinforcement Learning under a Pseudo Multi-Objective Reinforcement Learning Framework2022
- Author(s)
  Ilboudo Wendyam Eric Lionel, Kobayashi Taisuke, Matsubara Takamitsu
- Organizer
  日本ロボット学会学術講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] フェヒナーの法則に従う強化学習則の挙動解析2022
- Author(s)
  高橋慶一郎，小林泰介，松原崇充
- Organizer
  日本ロボット学会学術講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] 現方策による経験の到達可能性を考慮した強化学習2022
- Author(s)
  米澤壮太郎，小林泰介，松原崇充
- Organizer
  計測自動制御学会システムインテグレーション部門講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] Noise-Aware Stochastic Gradient Optimization with AdaTerm2022
- Author(s)
  Ilboudo Wendyam Eric Lionel, Kobayashi Taisuke, Matsubara Takamitsu
- Organizer
  IEEE/RSJ International Conference on Intelligent Robots and Systems
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] オフセット項付きフィードバック誤差学習による隊列走行の縦方向制御2022
- Author(s)
  中尾安宏、永田篤樹、小林泰介、杉本謙二
- Organizer
  第66回システム制御情報学会研究発表講演会
- Data Source
  KAKENHI-PROJECT-23K20948
[Presentation] Static output feedback synthesis of time-delay linear systems via deep unfolding2022
- Author(s)
  Ogura, M., Kobayashi, K., & Sugimoto, K.
- Organizer
  17th IFAC Workshop on Time Delay Systems
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-23K20948
[Presentation] Intentional Underestimation at Terminal State in Reinforcement Learning2022
- Author(s)
  Taisuke Kobayashi
- Organizer
  The 54th ISCIE International Symposium on Stochastic Systems Theory and Its Applications
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-23K20948
[Presentation] 強化学習における指数移動平均フィルタの統合2021
- Author(s)
  佐伯雄飛，小林泰介，杉本謙二
- Organizer
  日本ロボット学会学術講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] カルバック・ライブラ情報量に関する最適化問題としてのリスク回避型強化学習の提案2021
- Author(s)
  小林泰介
- Organizer
  日本ロボット学会学術講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] ツァリス統計に基づく変分オートエンコーダによるスパースな潜在空間の獲得2021
- Author(s)
  綿貫零真，小林泰介，杉本謙二
- Organizer
  日本ロボット学会学術講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] Towards Deep Robot Learning with Optimizer Applicable to Non-Stationary Problems2021
- Author(s)
  Kobayashi Taisuke
- Organizer
  IEEE/SICE International Symposium on System Integration
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] フィードバック誤差学習による隊列走行車両の縦方向制御2021
- Author(s)
  中尾安宏，小林泰介，杉本謙二
- Organizer
  第65回システム制御情報学会研究発表講演会
- Data Source
  KAKENHI-PROJECT-23K20948
[Presentation] 強化学習における局所リプシッツ連続に関する正則化2021
- Author(s)
  小林泰介
- Organizer
  自律分散システム・シンポジウム
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] 拡大Tchebyshev関数を用いた多目的最適化としての潜在動的モデルの学習2021
- Author(s)
  武田俊季，小林泰介，杉本謙二
- Organizer
  制御部門マルチシンポジウム
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] フィードバック・フィードフォワード方策を内包する強化学習アルゴリズム2021
- Author(s)
  小林泰介，芳澤健太
- Organizer
  第26回ロボティクスシンポジア
- Data Source
  KAKENHI-PROJECT-18H01462
[Presentation] Proximal Policy Optimization with Relative Pearson Divergence2021
- Author(s)
  Kobayashi Taisuke
- Organizer
  IEEE International Conference on Robotics and Automation
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] フィードバック誤差学習による隊列走行　―縦方向制御のロバスト化と外乱抑制―2021
- Author(s)
  中尾安宏，小林泰介，杉本謙二
- Organizer
  第64回自動制御連合講演会
- Data Source
  KAKENHI-PROJECT-23K20948
[Presentation] 潜在空間におけるモデル予測制御2020
- Author(s)
  武田敏季，小林泰介，杉本謙二
- Organizer
  ロボティクス・メカトロニクス講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] 紐解かれた潜在空間抽出のためのツァリス統計型変分オートエンコーダ2020
- Author(s)
  小林泰介
- Organizer
  ロボティクス・メカトロニクス講演会
- Data Source
  KAKENHI-PROJECT-20H04265
[Presentation] パラメータの定着とスパース化を統合した正則化による継続学習2019
- Author(s)
  小林泰介
- Organizer
  日本ロボット学会学術講演会
- Data Source
  KAKENHI-PROJECT-17K12759
[Presentation] Continual Learning Exploiting Structure of Fractal Reservoir Computing2019
- Author(s)
  Taisuke Kobayashi and Toshiki Sugino
- Organizer
  International Conference on Artificial Neural Networks
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-17K12759
[Presentation] フラクタルリザーバコンピューティングを用いた４脚ロボットの階層強化学習2019
- Author(s)
  杉野峻生，小林泰介，杉本謙二
- Organizer
  ロボティクス・メカトロニクス講演会
- Data Source
  KAKENHI-PROJECT-17K12759
[Presentation] Hyperbolically-Discounted Reinforcement Learning on Reward-Punishment Framework2019
- Author(s)
  Taisuke Kobayashi
- Organizer
  International Conference on Development and Learning and Epigenetic Robotics
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-18H01462
[Presentation] Check Regularization: Combining Modularity and Elasticity for Memory Consolidation2018
- Author(s)
  Taisuke Kobayashi
- Organizer
  International Conference on Artificial Neural Networks
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-17K12759
[Presentation] Continual Learning using Modularity of Structured Reservoir Computing2018
- Author(s)
  Toshiki Sugino, Taisuke Kobayashi, Kenji Sugimoto
- Organizer
  SICE Annual Conference
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-17K12759
[Presentation] フラクタルリザーバコンピューティングを用いた継続学習2018
- Author(s)
  杉野峻生，小林泰介，杉本謙二
- Organizer
  ロボティクス・メカトロニクス講演会
- Data Source
  KAKENHI-PROJECT-17K12759
[Presentation] 大域的最適解を目指すActor-Critic強化学習2017
- Author(s)
  小林泰介
- Organizer
  日本ロボット学会学術講演会
- Data Source
  KAKENHI-PROJECT-17K12759

1. Sugimoto Kenji (20179154)

# of Collaborated Projects: 2 results

# of Collaborated Products: 5 results
2. 小蔵正輝 (10800732)

# of Collaborated Projects: 2 results

# of Collaborated Products: 1 results
3. 橘拓至 (20415847)

# of Collaborated Projects: 2 results

# of Collaborated Products: 0 results
4. 潮俊光 (30184998)

# of Collaborated Projects: 2 results

# of Collaborated Products: 0 results
5. 花田研太 (80783703)

# of Collaborated Projects: 2 results

# of Collaborated Products: 0 results
6. 橋本和宗 (60883361)

# of Collaborated Projects: 2 results

# of Collaborated Products: 0 results
7. 坂本登 (00283416)

# of Collaborated Projects: 1 results

# of Collaborated Products: 0 results
8. 中島明 (70377836)

# of Collaborated Projects: 1 results

# of Collaborated Products: 0 results
9. 松原崇充 (20508056)

# of Collaborated Projects: 1 results

# of Collaborated Products: 0 results
10. 鶴峯義久 (00908113)

# of Collaborated Projects: 1 results

# of Collaborated Products: 0 results

Kobayashi Taisuke 小林 泰介

Research Projects

Research Products

Co-Researchers

Elucidating Functional Roles of Nonlinear TD Errors Guided by Reinforcement Learning as InferencePrincipal Investigator

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

Efficient and rational prediction process integrated with optimality-conditioned internal modelPrincipal Investigator

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

計測・通信品質が保証されない環境下の多目的フィードフォワード最適制御と強化学習

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

Development of deep learning to reveal physical human-robot interaction and its application to safe robot controlPrincipal Investigator

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

Event-trigger-type two-degree-of-freedom control system without quality assurance of measurement/communication

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

Locomotion Control for Legged Robot using Hierarchical Multi-Objective Reinforcement LearningPrincipal Investigator

Principal Investigator

Project Period (FY)

Research Category

Research Field

Research Institution

[Journal Article] Constrained footstep planning using model-based reinforcement learning in virtual constraint-based walking2024

Author(s)

Journal Title

DOI

Data Source

[Journal Article] Sparse representation learning with modified q-VAE towards minimal realization of world model2023

Author(s)

Journal Title

DOI

Data Source

[Journal Article] AdaTerm: Adaptive T-distribution estimated robust moments for Noise-Robust stochastic gradient optimization2023

Author(s)

Journal Title

DOI

Data Source

[Journal Article] Reward bonuses with gain scheduling inspired by iterative deepening search2023

Author(s)

Journal Title

DOI

Data Source

[Journal Article] Proximal policy optimization with adaptive threshold for symmetric relative density ratio2023

Author(s)

Journal Title

DOI

Data Source

[Journal Article] Longitudinal Control for Platooning of Vehicles by Two-degree-of-freedom Structure with Online Tuning2022

Author(s)

Journal Title

DOI

ISSN

Language

Data Source

[Journal Article] オンライン調整を伴う2自由度構成による隊列走行車両の縦方向制御2022

Author(s)

Journal Title

Data Source

[Journal Article] Adaptive and multiple time-scale eligibility traces for online deep reinforcement learning2022

Author(s)

Journal Title

DOI

Kobayashi Taisuke 小林泰介