Research map / Reinforcement learning / Policy gradient and actor-critic

Actor Prioritized Experience Replay (Abstract Reprint)

AAAI 2024 · Reinforcement learning › Policy gradient and actor-critic · approach: Stabilizing deep actor-critic · 2 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Studies Prioritized Experience Replay (PER) for deep RL and proposes an actor-prioritized variant of replay, based only on the abstract (full text unavailable).

Problem

Prioritized Experience Replay samples transitions by TD error in deep RL; the abstract indicates its benefit is limited in some settings (truncated), so the paper targets PER's effectiveness in continuous action domains.

Approach

Based on the truncated abstract, the paper revisits PER for continuous-control actor-critic methods and proposes prioritizing actor-related experience replay; the precise mechanism is not stated in the available text.

Limitations

• Full text unavailable; only the truncated abstract was available

Experimental setup

Builds on: Prioritized Experience Replay

Closest papers by meaning

• Sample Efficient Actor-Critic with Experience Replay (ICLR 2017 · similarity 0.92)
• An Equivalence between Loss Functions and Non-Uniform Sampling in Experience Replay (NeurIPS 2020 · similarity 0.91)
• Large Batch Experience Replay (ICML 2022 · similarity 0.91)
• Model-augmented Prioritized Experience Replay (ICLR 2022 · similarity 0.91)
• Off-Policy Actor-Critic with Shared Experience Replay (ICML 2020 · similarity 0.91)
• Prioritized Generative Replay (ICLR 2025 · similarity 0.89)