mlsanityData to 14 Oct 2026

Pessimistic Minimax Learning for Public-Private Information Games under Unilateral Coverage

arXiv 2610.04997 ↗First posted 4 Oct 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

We study offline learning in two-player zero-sum contextual games with public and private information, motivated by strategic settings such as auctions and negotiations with private valuations. We introduce unilateral prescriptive concentrability and show that asymmetric information can change offline coverage through its effect on equilibrium behavior. For finite state-action spaces, we develop a pessimistic algorithm with an tildeO(1/√n) exploitability rate, matching the standard sample-size dependence for fully observed minimax games. We further develop a pessimistic policy mirror descent framework, PPA-PMD, for general function approximation and obtain a unified tildeO(1/√n + 1/√T) exploitability rate with no-regret actor updates. Together, these results provide the first theoretical framework for offline equilibrium learning under public-private information constraints.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 35 of its 57 references that are in the map. First posted on arXiv 4 Oct 2026.

Builds on

35 of 57 references.

References · 35

  1. 2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemscited by 1,920
  2. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,809
  3. 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,558
  4. 2018Off-Policy Deep Reinforcement Learning without Explorationcited by 1,241
  5. 2017Counterfactual Multi-Agent Policy Gradientscited by 751
  6. 2019Behavior Regularized Offline Reinforcement Learningcited by 662
  7. 2020Is Pessimism Provably Efficient for Offline RL?cited by 364
  8. 2021Bridging Offline Reinforcement Learning and Imitation Learning: A Tale of Pessimismcited by 279
  9. 2021Bellman-consistent Pessimism for Offline Reinforcement Learningcited by 269
  10. 2021Pessimistic Model-based Offline Reinforcement Learning under Partial Coveragecited by 164
  11. 2020Batch Value-function Approximation with Only Realizabilitycited by 104
  12. 2022Near-optimal Offline Reinforcement Learning with Linear Representation: Leveraging Variance Information with Pessimismcited by 65
23 more
  1. 2021Plan Better Amid Conservatism: Offline Multi-Agent Reinforcement Learning with Actor Rectificationcited by 61
  2. 2021Contrasting Centralized and Decentralized Critics in Multi-Agent Reinforcement Learningcited by 56
  3. 2022Pessimistic Minimax Value Iteration: Provably Efficient Equilibrium Learning from Offline Datasetscited by 44
  4. 2022Provably Efficient Offline Multi-agent Reinforcement Learning via Strategy-wise Bonuscited by 30
  5. 2024The ODE Method for Stochastic Approximation and Reinforcement Learning with Markovian Noisecited by 24
  6. 2023Offline Learning in Markov Games with General Function Approximationcited by 18
  7. 2023Efficient Policy Evaluation with Offline Data Informed Behavior Policy Designcited by 14
  8. 2024Sample Complexity of Posted Pricing for a Single Itemcited by 11
  9. 2026Beyond Pessimism: Offline Learning in KL-regularized Gamescited by 10
  10. 2024Efficient Multi-Policy Evaluation for Reinforcement Learningcited by 9
  11. 2023Off-the-Grid MARL: Datasets with Baselines for Offline Multi-Agent Reinforcement Learningcited by 8
  12. 2024Doubly Optimal Policy Evaluation for Reinforcement Learningcited by 8
  13. 2026Instructing LLMs to Negotiate using Reinforcement Learning with Verifiable Rewardscited by 7
  14. 2024Efficient Policy Evaluation with Safety Constraint for Reinforcement Learningcited by 6
  15. 2026Offline Two-Player Zero-Sum Markov Games with KL Regularizationcited by 6
  16. 2026Pessimism-Free Offline Learning in General-Sum Games via KL Regularizationcited by 5
  17. 2026Fast Rates in α-Potential Games via Regularized Mirror Descentcited by 5
  18. 2026Beyond Linear Attention: Softmax Transformers Implement In-Context Reinforcement Learningcited by 5
  19. 2026Strategic Bargaining in Multi-Buyer Markets: Reinforcement Learning from Verifiable Rewards for LLM Negotiationscited by 4
  20. 2026Convergence of Two-Timescale Markovian Stochastic Approximations with Applications in Reinforcement Learningcited by 3
  21. 2026Robust Data-Collection Policy Learning for Low-Variance Online Policy Evaluationcited by 2
  22. 2026OR-Transformer: Scaling Real-Time Decision-Making to 1,000 Itemscited by 1
  23. 2026Learning to Sell: Reinforcement Learning for Strategic Large Language Model Agents in Multi-Product Marketscited by 1

Built on by

Not cited yet.