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

CHPO: Constrained Hybrid-action Policy Optimization for Reinforcement Learning

NeurIPS 2025 · Reinforcement learning › Policy gradient and actor-critic · approach: Constrained policy optimization · 1 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.

Proposes CHPO, a constrained RL method for discrete-continuous hybrid action spaces that aims to maximize reward while satisfying cost constraints.

Problem

Existing hybrid-action RL algorithms focus on reward maximization and struggle when tasks also impose cost constraints, which matters for safety-critical applications with discrete-continuous action spaces.

Approach

Constrained policy optimization tailored to parameterized hybrid action spaces, learning a safe policy within the hybrid action space; the abstract is truncated so the specific mechanism is not stated.

Closest papers by meaning

• Constraint-Conditioned Policy Optimization for Versatile Safe Reinforcement Learning (NeurIPS 2023 · similarity 0.89)
• Efficient Action-Constrained Reinforcement Learning via Acceptance-Rejection Method and Augmented MDPs (ICLR 2025 · similarity 0.87)
• Constrained Reinforcement Learning Under Model Mismatch (ICML 2024 · similarity 0.87)
• Hybrid Reinforcement Learning Breaks Sample Size Barriers In Linear MDPs (NeurIPS 2024 · similarity 0.87)
• Last-Iterate Global Convergence of Policy Gradients for Constrained Reinforcement Learning (NeurIPS 2024 · similarity 0.87)
• ACPO: A Policy Optimization Algorithm for Average MDPs with Constraints (ICML 2024 · similarity 0.86)