Pith. sign in

REVIEW 3 cited by

A general class of surrogate functions for stable and efficient reinforcement learning

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 2108.05828 v5 pith:X3O5THQB submitted 2021-08-12 cs.LG cs.AIstat.ML

classification cs.LGcs.AIstat.ML
keywords surrogatefunctionsfma-pgguaranteespolicyalgorithmsframeworkgeneral
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Common policy gradient methods rely on the maximization of a sequence of surrogate functions. In recent years, many such surrogate functions have been proposed, most without strong theoretical guarantees, leading to algorithms such as TRPO, PPO or MPO. Rather than design yet another surrogate function, we instead propose a general framework (FMA-PG) based on functional mirror ascent that gives rise to an entire family of surrogate functions. We construct surrogate functions that enable policy improvement guarantees, a property not shared by most existing surrogate functions. Crucially, these guarantees hold regardless of the choice of policy parameterization. Moreover, a particular instantiation of FMA-PG recovers important implementation heuristics (e.g., using forward vs reverse KL divergence) resulting in a variant of TRPO with additional desirable properties. Via experiments on simple bandit problems, we evaluate the algorithms instantiated by FMA-PG. The proposed framework also suggests an improved variant of PPO, whose robustness and efficiency we empirically demonstrate on the MuJoCo suite.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

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

  1. On the Policy Convergence of Policy Mirror Descent Methods

    math.OC 2026-07 accept novelty 7.0 of 10

    Unregularized PMD with any constant step size converges to a limiting optimal policy for general decomposable Legendre mirror maps, with behavior governed by differentiability of ψ at 0 and 1.

  2. Fast Convergence of Softmax Policy Mirror Ascent

    cs.LG 2024-11 conditional novelty 6.0 of 10

    Softmax policy mirror ascent is a normalization-free mirror ascent on logits that converges linearly in tabular MDPs and linearly to a neighborhood with function approximation.

  3. Multi-Agent Reinforcement Learning in Wireless Distributed Networks for 6G

    cs.IT 2025-02 conditional novelty 1.0 of 10

    A comprehensive survey of multi-agent reinforcement learning for wireless distributed networks in 6G, covering structures, algorithms, enhanced techniques, and applications.

Pith tools