Pith. sign in

REVIEW 2 cited by

Policy Gradient in Partially Observable Environments: Approximation and Convergence

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 1810.07900 v3 pith:DKXN2S3I submitted 2018-10-18 cs.LG stat.ML

classification cs.LGstat.ML
keywords observablegradientpolicypartiallyconvergenceenvironmentsapproachfully
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Policy gradient is a generic and flexible reinforcement learning approach that generally enjoys simplicity in analysis, implementation, and deployment. In the last few decades, this approach has been extensively advanced for fully observable environments. In this paper, we generalize a variety of these advances to partially observable settings, and similar to the fully observable case, we keep our focus on the class of Markovian policies. We propose a series of technical tools, including a novel notion of advantage function, to develop policy gradient algorithms and study their convergence properties in such environments. Deploying these tools, we generalize a variety of existing theoretical guarantees, such as policy gradient and convergence theorems, to partially observable domains, those which also could be carried to more settings of interest. This study also sheds light on the understanding of policy gradient approaches in real-world applications which tend to be partially observable.

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

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

  1. Learning Deterministic Policies with Policy Gradients in Constrained Markov Decision Processes

    cs.LG 2025-06 conditional novelty 6.0 of 10

    A primal-dual policy-gradient method with ridge regularization is shown to converge globally, in the last iterate, to optimal feasible deterministic policies in continuous constrained MDPs under gradient-domination an...

  2. Evolutionary Optimization of Deep Learning Agents for Sparrow Mahjong

    cs.NE 2025-08 unverdicted novelty 4.0 of 10

    A CMA-ES-optimized LSTM agent for Sparrow Mahjong is claimed to beat random and rule-based agents and match a PPO baseline, but the provided manuscript contains no verifiable experiments.

Pith tools