Ariu Kaito 蟻生開人

Researcher Number	80984286
Other IDs
Affiliation (Current)	2026: 株式会社サイバーエージェント(AI事業本部 AI Lab), AItech Studio AI Lab, リサーチサイエンティスト
Affiliation (based on the past Project Information) *help	2024 – 2025: 株式会社サイバーエージェント(AI事業本部 AI Lab), AItech Studio AI Lab, リサーチサイエンティスト(上席) 2023: 株式会社サイバーエージェント(AI事業本部 AI Lab), AItech Studio AI Lab, リサーチサイエンティスト
Review Section/Research Field	Principal Investigator Basic Section 61030:Intelligent informatics-related / 1001:Information science, computer engineering, and related fields
Keywords	Principal Investigator オンライン学習 / 逐次的意思決定 / 多腕バンディット問題 / 最適腕識別 / Learning in Games / クラスタリング

Research Projects
(2 results)
Research Products
(31 results)

ミニマックス最適化で拓く最適方策探索手法の構築Principal Investigator
- Principal Investigator
  
  蟻生開人
- Project Period (FY)
  2025 – 2029
- Research Category
  
  Grant-in-Aid for Early-Career Scientists
- Review Section
  
  Basic Section 61030:Intelligent informatics-related
- Research Institution
  CyberAgent, Inc. AI tech studio AI Lab
Construction of Large-Scale Sequential Decision-Making Methods Leveraging StructuresPrincipal Investigator
- Principal Investigator
  
  Ariu Kaito
- Project Period (FY)
  2023 – 2024
- Research Category
  
  Grant-in-Aid for Research Activity Start-up
- Review Section
  
  1001:Information science, computer engineering, and related fields
- Research Institution
  CyberAgent, Inc. AI tech studio AI Lab

All 2025 2024 2023

All Journal Article Presentation

[Journal Article] Global Behavior of Learning Dynamics in Zero-Sum Games with Memory Asymmetry2025
- Author(s)
  Yuma Fujimoto, Kaito Ariu, Kenshi Abe
- Journal Title
  
  Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS)
  
  Volume: -
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Theoretical Guarantees for Minimum Bayes Risk Decoding2025
- Author(s)
  Yuki Ichihara, Yuu Jinnai, Kaito Ariu, Tetsuro Morimura, Eiji Uchibe
- Journal Title
  
  Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL)
  
  Volume: -
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Return-Aligned Decision Transformer2025
- Author(s)
  Tsunehiko Tanaka, Kenshi Abe, Kaito Ariu, Tetsuro Morimura, Edgar Simo-Serra
- Journal Title
  
  Transactions on Machine Learning Research
  
  Volume: -
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Boosting Perturbed Gradient Ascent for Last-Iterate Convergence in Games2025
- Author(s)
  Kenshi Abe, Mitsuki Sakamoto, Kaito Ariu, Atsushi Iwasaki
- Journal Title
  
  The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025
  
  Volume: -
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Efficient Creative Selection in Online Advertising using Top-Two Thompson Sampling2025
- Author(s)
  Daiki Katsuragawa, Yusuke Kaneko, Kaito Ariu, Kenshi Abe
- Journal Title
  
  In Proceedings of the Eighteenth ACM International Conference on Web Search and Data Mining (WSDM '25)
  
  Volume: - Pages: 1090-1091
- DOI
  10.1145/3701551.3706128
- Peer Reviewed
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Nash Equilibrium and Learning Dynamics in Three-Player Matching m-Action Games2025
- Author(s)
  Yuma Fujimoto, Kaito Ariu, Kenshi Abe
- Journal Title
  
  Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Extended Abstract
  
  Volume: -
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Revisiting Instance-Optimal Cluster Recovery in the Labeled Stochastic Block Model2025
- Author(s)
  Kaito Ariu, Alexandre Proutiere, Se-Young Yun
- Journal Title
  
  Proceedings of the 42nd International Conference on Machine Learning (ICML 2025)
  
  Volume: -
- Peer Reviewed / Open Access / Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Evaluation of Best-of-N Sampling Strategies for Language Model Alignment2025
- Author(s)
  Yuki Ichihara, Yuu Jinnai, Tetsuro Morimura, Kenshi Abe, Kaito Ariu, Mitsuki Sakamoto, Eiji Uchibe
- Journal Title
  
  Transactions on Machine Learning Research
  
  Volume: -
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Synchronization in Learning in Periodic Zero-Sum Games Triggers Divergence from Nash Equilibrium2025
- Author(s)
  Yuma Fujimoto, Kaito Ariu, Kenshi Abe
- Journal Title
  
  Proceedings of the AAAI Conference on Artificial Intelligence
  
  Volume: 39 Issue: 22 Pages: 23194-23202
- DOI
  10.1609/aaai.v39i22.34485
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Regularized Best-of-N Sampling with Minimum Bayes Risk Objective for Language Model Alignment2025
- Author(s)
  Yuu Jinnai, Tetsuro Morimura, Kaito Ariu, Kenshi Abe
- Journal Title
  
  Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
  
  Volume: -
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Hyperparameter-Free Approach for Faster Minimum Bayes Risk Decoding2024
- Author(s)
  Yuu Jinnai, Kaito Ariu
- Journal Title
  
  Findings of the Association for Computational Linguistics: ACL 2024
  
  Volume: - Pages: 8547-8566
- DOI
  10.18653/v1/2024.findings-acl.505
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Matroid semi-bandits in sublinear time2024
- Author(s)
  Ruo-Chun Tzeng, Naoto Ohsaka, Kaito Ariu
- Journal Title
  
  Proceedings of the 41st International Conference on Machine Learning (ICML 2024)
  
  Volume: 235
- Peer Reviewed / Open Access / Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Model-based minimum Bayes risk decoding for text generation2024
- Author(s)
  Yuu Jinnai, Tetsuro Morimura, Ukyo Honda, Kaito Ariu, Kenshi Abe
- Journal Title
  
  Proceedings of the 41st International Conference on Machine Learning (ICML 2024)
  
  Volume: 235
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] On universally optimal algorithms for A/B testing2024
- Author(s)
  Po-An Wang, Kaito Ariu, Alexandre Proutiere
- Journal Title
  
  Proceedings of the 41st International Conference on Machine Learning (ICML 2024)
  
  Volume: 235
- Peer Reviewed / Open Access / Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Optimal clustering from noisy binary feedback2024
- Author(s)
  Ariu Kaito、Ok Jungseul、Proutiere Alexandre、Yun Seyoung
- Journal Title
  
  Machine Learning
  
  Volume: 113 Issue: 5 Pages: 2733-2764
- DOI
  10.1007/s10994-024-06532-z
- Peer Reviewed / Open Access / Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Adaptively perturbed mirror descent for learning in games2024
- Author(s)
  Kenshi Abe, Kaito Ariu, Mitsuki Sakamoto, Atsushi Iwasaki
- Journal Title
  
  Proceedings of the 41st International Conference on Machine Learning (ICML 2024)
  
  Volume: 235
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986
[Journal Article] Memory Asymmetry Creates Heteroclinic Orbits to Nash Equilibrium in Learning in Zero-Sum Games2024
- Author(s)
  Fujimoto Yuma、Ariu Kaito、Abe Kenshi
- Journal Title
  
  Proceedings of the AAAI Conference on Artificial Intelligence
  
  Volume: 38 Issue: 16 Pages: 17398-17406
- DOI
  10.1609/aaai.v38i16.29688
- Peer Reviewed / Open Access
- Data Source
  KAKENHI-PROJECT-23K19986, KAKENHI-PROJECT-22KJ1414
[Presentation] 大規模言語モデルのためのアライメントデータ合成手法の実験的評価2025
- Author(s)
  坂本充生, 陣内佑, 森村哲郎, 阿部拳之, 蟻生開人
- Organizer
  言語処理学会第31回年次大会(NLP2025)
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] テキスト生成における最小ベイズリスク復号の理論的な理解に向けて2025
- Author(s)
  市原有生希, 陣内佑, 蟻生開人, 森村哲郎, 内部英治
- Organizer
  言語処理学会第31回年次大会(NLP2025)
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] Synchronization behind Learning in Periodic Zero-Sum Games Triggers Divergence from Nash equilibrium2024
- Author(s)
  藤本悠雅, 蟻生開人, 阿部拳之
- Organizer
  第27回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] マルコフ決定過程における良方策検定手法の提案2024
- Author(s)
  蟻生開人, Po-An Wang, 阿部拳之, Alexandre Proutiere
- Organizer
  第27回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] Evaluation of Best-of-N Sampling Strategies for Language Model Alignment2024
- Author(s)
  市原有生希, 陣内佑, 森村哲郎, 阿部拳之, 蟻生開人, 坂本充生, 内部英治
- Organizer
  第27回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] （不完全情報）展開型ゲームにおける零分散の利得摂動手法2024
- Author(s)
  眞坂航宙, 坂本充生, 阿部拳之, 蟻生開人, 岩崎敦
- Organizer
  第27回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] ベイズリスク選好最適化：報酬モデル不要のオンライン選好最適化手法2024
- Author(s)
  森村哲郎, 坂本充生, 陣内佑, 阿部拳之, 蟻生開人
- Organizer
  第27回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] Last Iterate Convergence in Monotone Mean Field Games2024
- Author(s)
  磯部伸, 阿部拳之, 蟻生開人
- Organizer
  第27回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] Filtered Direct Preference Optimization: 選好データセットの質に基づくフィルタリング手法の提案2024
- Author(s)
  坂本充生, 森村哲郎, 陣内佑, 阿部拳之, 蟻生開人
- Organizer
  第27回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] A Slingshot Approach to Learning in Monotone Games2023
- Author(s)
  阿部拳之、蟻生開人、坂本充生、岩崎敦
- Organizer
  第26回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] On Uniformly Optimal Algorithms for Best-Arm Identification in Two-Armed Bandits with Fixed Budget2023
- Author(s)
  Wang Po-An、Ariu Kaito、Proutiere Alexandre
- Organizer
  Workshop Tests and Bandits in Potsdam 2023
- Int'l Joint Research
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] Leveraging Perturbation for Convergence in Extensive-Form Games2023
- Author(s)
  坂本充生、阿部拳之、蟻生開人、岩崎敦
- Organizer
  第26回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] 固定予算二腕最適腕識別における一様サンプリングの最適性について2023
- Author(s)
  蟻生開人、Wang Po-An、Proutiere Alexandre
- Organizer
  第26回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986
[Presentation] Learning in Multi-Memory Games Triggers Complex Dynamics Diverging from Nash equilibrium2023
- Author(s)
  藤本悠雅、蟻生開人、阿部拳之
- Organizer
  第26回情報論的学習理論ワークショップ
- Data Source
  KAKENHI-PROJECT-23K19986

Ariu Kaito 蟻生 開人

Research Projects

Research Products

ミニマックス最適化で拓く最適方策探索手法の構築Principal Investigator

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

Construction of Large-Scale Sequential Decision-Making Methods Leveraging StructuresPrincipal Investigator

Principal Investigator

Project Period (FY)

Research Category

Review Section

Research Institution

[Journal Article] Global Behavior of Learning Dynamics in Zero-Sum Games with Memory Asymmetry2025

Author(s)

Journal Title

Data Source

[Journal Article] Theoretical Guarantees for Minimum Bayes Risk Decoding2025

Author(s)

Journal Title

Data Source

[Journal Article] Return-Aligned Decision Transformer2025

Author(s)

Journal Title

Data Source

[Journal Article] Boosting Perturbed Gradient Ascent for Last-Iterate Convergence in Games2025

Author(s)

Journal Title

Data Source

[Journal Article] Efficient Creative Selection in Online Advertising using Top-Two Thompson Sampling2025

Author(s)

Journal Title

DOI

Data Source

[Journal Article] Nash Equilibrium and Learning Dynamics in Three-Player Matching m-Action Games2025

Author(s)

Journal Title

Data Source

[Journal Article] Revisiting Instance-Optimal Cluster Recovery in the Labeled Stochastic Block Model2025

Author(s)

Journal Title

Data Source

[Journal Article] Evaluation of Best-of-N Sampling Strategies for Language Model Alignment2025

Author(s)

Journal Title

Data Source

[Journal Article] Synchronization in Learning in Periodic Zero-Sum Games Triggers Divergence from Nash Equilibrium2025

Author(s)

Journal Title

DOI

Data Source

[Journal Article] Regularized Best-of-N Sampling with Minimum Bayes Risk Objective for Language Model Alignment2025

Author(s)

Journal Title

Data Source

[Journal Article] Hyperparameter-Free Approach for Faster Minimum Bayes Risk Decoding2024

Author(s)

Journal Title

DOI

Data Source

[Journal Article] Matroid semi-bandits in sublinear time2024

Author(s)

Journal Title

Data Source

[Journal Article] Model-based minimum Bayes risk decoding for text generation2024

Author(s)

Journal Title

Data Source

[Journal Article] On universally optimal algorithms for A/B testing2024

Author(s)

Journal Title

Data Source

[Journal Article] Optimal clustering from noisy binary feedback2024

Author(s)

Journal Title

DOI

Data Source

[Journal Article] Adaptively perturbed mirror descent for learning in games2024

Ariu Kaito 蟻生開人