mlsanityData to 7 Oct 2026

Specification-Guided Learning of Nash Equilibria with High Social Welfare

arXiv 2206.03348 ↗First posted 6 Jun 2022cs.GT cs.AI cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

Reinforcement learning has been shown to be an effective strategy for automatically training policies for challenging control problems. Focusing on non-cooperative multi-agent systems, we propose a novel reinforcement learning framework for training joint policies that form a Nash equilibrium. In our approach, rather than providing low-level reward functions, the user provides high-level specifications that encode the objective of each agent. Then, guided by the structure of the specifications, our algorithm searches over policies to identify one that provably forms an ε-Nash equilibrium (with high probability). Importantly, it prioritizes policies in a way that maximizes social welfare across all agents. Our empirical evaluation demonstrates that our algorithm computes equilibrium policies with high social welfare, whereas state-of-the-art baselines either fail to compute Nash equilibria or compute ones with comparatively lower social welfare.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 20 of its 51 references that are in the map. First posted on arXiv 6 Jun 2022.

Builds on

20 of 51 references.

References · 20

  1. 2020Reward Machines: Exploiting Reward Function Structure in Reinforcement Learningcited by 183
  2. 2020Reward-Free Exploration for Reinforcement Learningcited by 177
  3. 2016Reinforcement Learning With Temporal Logic Rewardscited by 114
  4. 2020Provable Self-Play Algorithms for Competitive Reinforcement Learningcited by 112
  5. 2019Reinforcement Learning for Temporal Logic Control Synthesis with Probabilistic Satisfaction Guaranteescited by 77
  6. 2018Omega-Regular Objectives in Model-Free Reinforcement Learningcited by 74
  7. 2018Foundations for Restraining Bolts: Reinforcement Learning with LTLf/LDLf restraining specificationscited by 69
  8. 2020A Composable Specification Language for Reinforcement Learning Taskscited by 68
  9. 2017Environment-Independent Task Specifications via GLTLcited by 67
  10. 2017Online Reinforcement Learning in Stochastic Gamescited by 67
  11. 2016Q-Learning for Robust Satisfaction of Signal Temporal Logic Specificationscited by 65
  12. 2021Compositional Reinforcement Learning from Logical Specificationscited by 65
8 more
  1. 2018Logically-Constrained Reinforcement Learningcited by 50
  2. 2020Reward Machines for Cooperative Multi-Agent Reinforcement Learningcited by 35
  3. 2019Transfer of Temporal Logic Formulas in Reinforcement Learningcited by 31
  4. 2021Multi-Agent Reinforcement Learning with Temporal Logic Specificationscited by 27
  5. 2019Modular Deep Reinforcement Learning with Temporal Logic Specificationscited by 25
  6. 2016Learning Nash Equilibrium for General-Sum Markov Games from Batch Datacited by 20
  7. 2018Equilibria-based Probabilistic Model Checking for Concurrent Stochastic Gamescited by 8
  8. 2013Fast Planning in Stochastic Gamescited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 1; 2024: 1; 2025: 3; 2026: 2.

8 papers.

  1. 2022COOL-MC: A Comprehensive Tool for Reinforcement Learning and Model Checkingcited by 19
  2. 2023Policy Synthesis and Reinforcement Learning for Discounted LTLcited by 14
  3. 2024Inductive Generalization in Reinforcement Learning from Specificationscited by 3
  4. 2025Turn-based Multi-Agent Reinforcement Learning Model Checkingcited by 3
  5. 2025HypRL: Reinforcement Learning of Control Policies for Hyperpropertiescited by 2
  6. 2025Large-Scale Auto-bidding with Nash Equilibrium Constraintscited by 0
  7. 2026Decoupled Behavioral Cloning for Scalable Inductive Generalization in RL from Specificationscited by 0
  8. 2026HyPOLE: Hyperproperty-Guided Multi-Agent Reinforcement Learning under Partial Observationcited by 0