Pith. sign in

REVIEW 2 cited by

Model-Free Learning for Two-Player Zero-Sum Partially Observable Markov Games with Perfect Recall

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2106.06279 v1 pith:VZI4VXVQ submitted 2021-06-11 stat.ML cs.LG

Model-Free Learning for Two-Player Zero-Sum Partially Observable Markov Games with Perfect Recall

classification stat.ML cs.LG
keywords gamelearningonlyalgorithmfeedbackgamesixomdmodel-free
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We study the problem of learning a Nash equilibrium (NE) in an imperfect information game (IIG) through self-play. Precisely, we focus on two-player, zero-sum, episodic, tabular IIG under the perfect-recall assumption where the only feedback is realizations of the game (bandit feedback). In particular, the dynamic of the IIG is not known -- we can only access it by sampling or interacting with a game simulator. For this learning setting, we provide the Implicit Exploration Online Mirror Descent (IXOMD) algorithm. It is a model-free algorithm with a high-probability bound on the convergence rate to the NE of order $1/\sqrt{T}$ where $T$ is the number of played games. Moreover, IXOMD is computationally efficient as it needs to perform the updates only along the sampled trajectory.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Differential Privacy in the Extensive-Form Bandit Problem

    cs.CR 2026-05 unverdicted novelty 7.0

    An algorithm achieves Õ(√(A ln(S) T)/ε) regret for extensive-form bandits under ε-local differential privacy, claimed as the first such result.

  2. Reinforcement Learning: From Algorithms To Foundation Models

    cs.AI 2026-07 conditional novelty 3.0

    A dissertation uniting the author's published results: non-exploitable Nash-DQN policies and the FightLadder benchmark for games, plus diffusion/consistency-model world models for RL — a compilation rather than new results.