Residual Policy Gradient: A Reward View of KL-regularized Objective

• URL: http://arxiv.org/abs/2503.11019v1
• Date: Fri, 14 Mar 2025 02:30:13 GMT
• Title: Residual Policy Gradient: A Reward View of KL-regularized Objective
• Authors: Pengcheng Wang, Xinghao Zhu, Yuxin Chen, Chenfeng Xu, Masayoshi Tomizuka, Chenran Li,
• Abstract summary: Reinforcement Learning and Imitation Learning have achieved widespread success in many domains but remain constrained during real-world deployment.<n>Policy customization has been introduced, aiming to adapt a prior policy while preserving its inherent properties and meeting new task-specific requirements.<n>A principled approach to policy customization is Residual Q-Learning (RQL), which formulates the problem as a Markov Decision Process (MDP) and derives a family of value-based learning algorithms.<n>We introduce Residual Policy Gradient (RPG), which extends RQL to policy gradient methods, allowing policy customization in gradient-based RL settings.
• Score: 48.39829592175419
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reinforcement Learning and Imitation Learning have achieved widespread success in many domains but remain constrained during real-world deployment. One of the main issues is the additional requirements that were not considered during training. To address this challenge, policy customization has been introduced, aiming to adapt a prior policy while preserving its inherent properties and meeting new task-specific requirements. A principled approach to policy customization is Residual Q-Learning (RQL), which formulates the problem as a Markov Decision Process (MDP) and derives a family of value-based learning algorithms. However, RQL has not yet been applied to policy gradient methods, which restricts its applicability, especially in tasks where policy gradient has already proven more effective. In this work, we first derive a concise form of Soft Policy Gradient as a preliminary. Building on this, we introduce Residual Policy Gradient (RPG), which extends RQL to policy gradient methods, allowing policy customization in gradient-based RL settings. With the view of RPG, we rethink the KL-regularized objective widely used in RL fine-tuning. We show that under certain assumptions, KL-regularized objective leads to a maximum-entropy policy that balances the inherent properties and task-specific requirements on a reward-level. Our experiments in MuJoCo demonstrate the effectiveness of Soft Policy Gradient and Residual Policy Gradient.

Related papers

• Landscape of Policy Optimization for Finite Horizon MDPs with General State and Action [10.219627570276689]
  We develop a framework for a class of Markov Decision Processes with general state and spaces. We show that gradient methods converge to the globally optimal policy with a nonasymptomatic condition. Our result establishes first complexity for multi-period inventory systems.
  arXiv  Detail & Related papers  (2024-09-25T17:56:02Z)
• Last-Iterate Global Convergence of Policy Gradients for Constrained Reinforcement Learning [62.81324245896717]
  We introduce an exploration-agnostic algorithm, called C-PG, which exhibits global last-ite convergence guarantees under (weak) gradient domination assumptions. We numerically validate our algorithms on constrained control problems, and compare them with state-of-the-art baselines.
  arXiv  Detail & Related papers  (2024-07-15T14:54:57Z)
• Projected Off-Policy Q-Learning (POP-QL) for Stabilizing Offline Reinforcement Learning [57.83919813698673]
  Projected Off-Policy Q-Learning (POP-QL) is a novel actor-critic algorithm that simultaneously reweights off-policy samples and constrains the policy to prevent divergence and reduce value-approximation error. In our experiments, POP-QL not only shows competitive performance on standard benchmarks, but also out-performs competing methods in tasks where the data-collection policy is significantly sub-optimal.
  arXiv  Detail & Related papers  (2023-11-25T00:30:58Z)
• Residual Q-Learning: Offline and Online Policy Customization without Value [53.47311900133564]
  Imitation Learning (IL) is a widely used framework for learning imitative behavior from demonstrations. We formulate a new problem setting called policy customization. We propose a novel framework, Residual Q-learning, which can solve the formulated MDP by leveraging the prior policy.
  arXiv  Detail & Related papers  (2023-06-15T22:01:19Z)
• Policy Gradient for Rectangular Robust Markov Decision Processes [62.397882389472564]
  We introduce robust policy gradient (RPG), a policy-based method that efficiently solves rectangular robust Markov decision processes (MDPs) Our resulting RPG can be estimated from data with the same time complexity as its non-robust equivalent.
  arXiv  Detail & Related papers  (2023-01-31T12:40:50Z)
• Sigmoidally Preconditioned Off-policy Learning:a new exploration method for reinforcement learning [14.991913317341417]
  We focus on an off-policy Actor-Critic architecture, and propose a novel method, called Preconditioned Proximal Policy Optimization (P3O) P3O can control the high variance of importance sampling by applying a preconditioner to the Conservative Policy Iteration (CPI) objective. Results show that our P3O maximizes the CPI objective better than PPO during the training process.
  arXiv  Detail & Related papers  (2022-05-20T09:38:04Z)
• Variance Reduction based Partial Trajectory Reuse to Accelerate Policy Gradient Optimization [3.621753051212441]
  We extend the idea of green simulation assisted policy gradient (GS-PG) to partial historical trajectory reuse for Markov Decision Processes (MDP) In this paper, the mixture likelihood ratio (MLR) based policy gradient estimation is used to leverage the information from historical state decision transitions generated under different behavioral policies.
  arXiv  Detail & Related papers  (2022-05-06T01:42:28Z)
• Towards an Understanding of Default Policies in Multitask Policy Optimization [29.806071693039655]
  Much of the recent success of deep reinforcement learning has been driven by regularized policy optimization (RPO) algorithms. We take a first step towards filling this gap by formally linking the quality of the default policy to its effect on optimization. We then derive a principled RPO algorithm for multitask learning with strong performance guarantees.
  arXiv  Detail & Related papers  (2021-11-04T16:45:15Z)
• Privacy-Constrained Policies via Mutual Information Regularized Policy Gradients [54.98496284653234]
  We consider the task of training a policy that maximizes reward while minimizing disclosure of certain sensitive state variables through the actions. We solve this problem by introducing a regularizer based on the mutual information between the sensitive state and the actions. We develop a model-based estimator for optimization of privacy-constrained policies.
  arXiv  Detail & Related papers  (2020-12-30T03:22:35Z)
• A Study of Policy Gradient on a Class of Exactly Solvable Models [35.90565839381652]
  We explore the evolution of the policy parameters, for a special class of exactly solvable POMDPs, as a continuous-state Markov chain. Our approach relies heavily on random walk theory, specifically on affine Weyl groups. We analyze the probabilistic convergence of policy gradient to different local maxima of the value function.
  arXiv  Detail & Related papers  (2020-11-03T17:27:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.