paper-with-me

Papers

A Simple Mixture Policy Parameterization for Improving Sample Efficiency of CVaR Optimization

2024-03-17 · Yudong Luo, Yangchen Pan, Han Wang, Philip Torr, Pascal Poupart

Reinforcement learning algorithms utilizing policy gradients (PG) to optimize Conditional Value at Risk (CVaR) face significant challenges with sample inefficiency, hindering their practical applications. This inefficiency stems from two main facts: a focus on tail-end performance that overlooks many sampled trajectories, and the potential of gradient vanishing when the lower tail of the return distribution is overly flat. To address these challenges, we propose a simple mixture policy parameterization. This method integrates a risk-neutral policy with an adjustable policy to form a risk-averse policy. By employing this strategy, all collected trajectories can be utilized for policy updating, and the issue of vanishing gradients is counteracted by stimulating higher returns through the risk-neutral component, thus lifting the tail and preventing flatness. Our empirical study reveals that this mixture parameterization is uniquely effective across a variety of benchmark domains. Specifically, it excels in identifying risk-averse CVaR policies in some Mujoco environments where the traditional CVaR-PG fails to learn a reasonable policy.

📄 PDF Abstract BibTeX arXiv:2403.11062

Code (0)

등록된 구현이 없습니다.

Tasks

MuJoCo

Methods 이 논문이 사용한 방법론

Focus 설명 없음

Similar Papers 제목 키워드 기반

Reparameterization Proximal Policy Optimization

2025-08-08 · Hai Zhong, Xun Wang, Zhuoran Li, Longbo Huang arxiv

By leveraging differentiable dynamics, Reparameterization Policy Gradient (RPG) achieves high sample efficiency. However, current approaches are hindered by two critical limitations: the under-utilization of computationa…

Reparameterization Flow Policy Optimization

2026-02-03 · Hai Zhong, Zhuoran Li, Xun Wang, Longbo Huang arxiv

Reparameterization Policy Gradient (RPG) has emerged as a powerful paradigm for model-based reinforcement learning, enabling high sample efficiency by backpropagating gradients through differentiable dynamics. However, p…

Reinforcement Learning

Soft Actor-Critic with Beta Policy via Implicit Reparameterization Gradients

2024-09-08 · Luca Della Libera

Recent advances in deep reinforcement learning have achieved impressive results in a wide range of complex tasks, but poor sample efficiency remains a major obstacle to real-world deployment. Soft actor-critic (SAC) miti…

continuous-controlContinuous ControlDeep Reinforcement Learning

QMP: Q-switch Mixture of Policies for Multi-Task Behavior Sharing

2023-02-01 · Grace Zhang, Ayush Jain, Injune Hwang, Shao-Hua Sun 외

Multi-task reinforcement learning (MTRL) aims to learn several tasks simultaneously for better sample efficiency than learning them separately. Traditional methods achieve this by sharing parameters or relabeled data bet…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)

Design of Restricted Normalizing Flow towards Arbitrary Stochastic Policy with Computational Efficiency

2024-12-17 · Taisuke Kobayashi, Takumi Aotani

This paper proposes a new design method for a stochastic control policy using a normalizing flow (NF). In reinforcement learning (RL), the policy is usually modeled as a distribution model with trainable parameters. When…

Computational EfficiencyReinforcement Learning (RL)