NashDreamer: Model-Based Reinforcement Learning for Zero-Sum Imperfect-Information Games

• URL: http://arxiv.org/abs/2609.01549v1
• Date: Tue, 01 Sep 2026 17:15:41 GMT
• Title: NashDreamer: Model-Based Reinforcement Learning for Zero-Sum Imperfect-Information Games
• Abstract summary: We propose a principled model-based reinforcement learning framework for imperfect information games (IIGs)<n>NashDreamer is designed to use arbitrary policy gradient algorithms and inherits their convergence guarantees towards Nash equilibria under an idealized model.<n> Empirical evaluations demonstrate that NashDreamer substantially improves sample efficiency over model-free baselines.
• Score: 3.4935179780034242
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Model-based reinforcement learning (MBRL) has achieved remarkable results in single-agent domains, yet its extension to competitive imperfect information games (IIGs) remains underexplored. In multi-agent settings, opponent-induced non-stationarity complicates the learning process, and decentralized model learning faces severe identifiability barriers, which we argue make centralized model learning a mathematical necessity. Building on this analysis, we propose NashDreamer, a principled MBRL framework for two-player zero-sum IIGs. It introduces a centralized Multi-Agent Recurrent State-Space Model (MARSSM) that decouples environment dynamics from the effect of players' strategies on their individual observations. NashDreamer is designed to use arbitrary policy gradient algorithms and inherits their convergence guarantees towards Nash equilibria under an idealized model. Empirical evaluations across four benchmark games demonstrate that NashDreamer substantially improves sample efficiency over model-free baselines early in the training. Finally, we theoretically analyze the architecture's optimization landscape, identifying the vulnerability of the Dreamer family of algorithms to posterior collapse in stochastic environments. We leave it as an open challenge.

Related papers

• Personalized and Resilient Distributed Learning Through Opinion Dynamics [1.1499574149885023]
  We address two practical challenges of distributed learning in multi-agent network systems, namely personalization and resilience.<n>Motivated by a conceptual affinity between these two requirements, we devise a distributed learning algorithm that combines distributed gradient descent and the Friedkin-Johnsen model of opinion dynamics.<n>We numerically showcase the effectiveness of our algorithm on synthetic and real-world distributed learning tasks, where it achieves high global accuracy both for personalized models and with malicious agents compared to standard strategies.
  arXiv  Detail & Related papers  (2025-05-20T08:39:16Z)
• Embodied-R: Collaborative Framework for Activating Embodied Spatial Reasoning in Foundation Models via Reinforcement Learning [58.86928947970342]
  Embodied-R is a framework combining large-scale Vision-Language Models for perception and small-scale Language Models for reasoning.<n>After training on only 5k embodied video samples, Embodied-R with a 3B LM matches state-of-the-art multimodal reasoning models.<n>Embodied-R also exhibits emergent thinking patterns such as systematic analysis and contextual integration.
  arXiv  Detail & Related papers  (2025-04-17T06:16:11Z)
• Double Oracle Neural Architecture Search for Game Theoretic Deep Learning Models [28.238075755838487]
  We propose a new approach to train deep learning models using game theory concepts. We deploy a double-versarial framework using best response oracles. We show that all our variants have significant improvements in both subjective qualitative evaluation and quantitative metrics.
  arXiv  Detail & Related papers  (2024-10-07T05:42:01Z)
• Preference-Based Multi-Agent Reinforcement Learning: Data Coverage and Algorithmic Techniques [65.55451717632317]
  We study Preference-Based Multi-Agent Reinforcement Learning (PbMARL)<n>We identify the Nash equilibrium from a preference-only offline dataset in general-sum games.<n>Our findings underscore the multifaceted approach required for PbMARL.
  arXiv  Detail & Related papers  (2024-09-01T13:14:41Z)
• Provably Efficient Information-Directed Sampling Algorithms for Multi-Agent Reinforcement Learning [50.92957910121088]
  This work designs and analyzes a novel set of algorithms for multi-agent reinforcement learning (MARL) based on the principle of information-directed sampling (IDS) For episodic two-player zero-sum MGs, we present three sample-efficient algorithms for learning Nash equilibrium. We extend Reg-MAIDS to multi-player general-sum MGs and prove that it can learn either the Nash equilibrium or coarse correlated equilibrium in a sample efficient manner.
  arXiv  Detail & Related papers  (2024-04-30T06:48:56Z)
• Model-Based RL for Mean-Field Games is not Statistically Harder than Single-Agent RL [57.745700271150454]
  We study the sample complexity of reinforcement learning in Mean-Field Games (MFGs) with model-based function approximation. We introduce the Partial Model-Based Eluder Dimension (P-MBED), a more effective notion to characterize the model class complexity.
  arXiv  Detail & Related papers  (2024-02-08T14:54:47Z)
• Deep autoregressive density nets vs neural ensembles for model-based offline reinforcement learning [2.9158689853305693]
  We consider a model-based reinforcement learning algorithm that infers the system dynamics from the available data and performs policy optimization on imaginary model rollouts. This approach is vulnerable to exploiting model errors which can lead to catastrophic failures on the real system. We show that better performance can be obtained with a single well-calibrated autoregressive model on the D4RL benchmark.
  arXiv  Detail & Related papers  (2024-02-05T10:18:15Z)
• HarmonyDream: Task Harmonization Inside World Models [93.07314830304193]
  Model-based reinforcement learning (MBRL) holds the promise of sample-efficient learning. We propose a simple yet effective approach, HarmonyDream, which automatically adjusts loss coefficients to maintain task harmonization.
  arXiv  Detail & Related papers  (2023-09-30T11:38:13Z)
• Efficient Model-based Multi-agent Reinforcement Learning via Optimistic Equilibrium Computation [93.52573037053449]
  H-MARL (Hallucinated Multi-Agent Reinforcement Learning) learns successful equilibrium policies after a few interactions with the environment. We demonstrate our approach experimentally on an autonomous driving simulation benchmark.
  arXiv  Detail & Related papers  (2022-03-14T17:24:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.