Policy Gradient Guidance Enables Test Time Control

• URL: http://arxiv.org/abs/2510.02148v1
• Date: Thu, 02 Oct 2025 16:00:35 GMT
• Title: Policy Gradient Guidance Enables Test Time Control
• Abstract summary: We introduce Policy Gradient Guidance (PGG)<n>PGG augments the policy gradient with an unconditional branch and interpolates conditional and unconditional branches.<n>We evaluate PGG on discrete and continuous control benchmarks.
• Score: 9.24766442685354
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We introduce Policy Gradient Guidance (PGG), a simple extension of classifier-free guidance from diffusion models to classical policy gradient methods. PGG augments the policy gradient with an unconditional branch and interpolates conditional and unconditional branches, yielding a test-time control knob that modulates behavior without retraining. We provide a theoretical derivation showing that the additional normalization term vanishes under advantage estimation, leading to a clean guided policy gradient update. Empirically, we evaluate PGG on discrete and continuous control benchmarks. We find that conditioning dropout-central to diffusion guidance-offers gains in simple discrete tasks and low sample regimes, but dropout destabilizes continuous control. Training with modestly larger guidance ($\gamma>1$) consistently improves stability, sample efficiency, and controllability. Our results show that guidance, previously confined to diffusion policies, can be adapted to standard on-policy methods, opening new directions for controllable online reinforcement learning.

Related papers

• A Unified and Constrained View of Regularization-Based Robust Reinforcement Learning [9.258167884740898]
  Regularization-based methods have become a standard approach for training policies against adversarial input perturbations.<n>In this paper, we unify these methods by deriving new upper bounds on the performance gap between the nominal and worst-case policies.<n>We also conduct extensive adversarial evaluations across several continuous control tasks to validate our theoretical analysis.
  arXiv  Detail & Related papers  (2026-09-11T16:45:36Z)
• DiPOD: Diffusion Policy Optimization without Drifting Apart [88.2714417395158]
  RL post-training has become increasingly pivotal for improving diffusion policies.<n>Existing diffusion policy-gradient methods are often unstable and cannot achieve reliable policy improvement.<n>We propose bftextDiPOD, a diffusion policy optimization framework that maintains tight-bound behavior throughout training.
  arXiv  Detail & Related papers  (2026-06-11T18:06:04Z)
• One-Way Policy Optimization for Self-Evolving LLMs [63.8638342097375]
  Reinforcement Learning with Verifiable Rewards (RLVR) has become a promising paradigm for scaling reasoning capabilities of Large Language Models (LLMs)<n>We propose One-Way Policy Optimization (OWPO), a method based on the principle of decoupling optimization direction from update magnitude.<n> Experimental results demonstrate that OWPO outperforms strong baselines, including DAPO, OPD, and MOPD.
  arXiv  Detail & Related papers  (2026-05-21T08:25:27Z)
• OGPO: Sample Efficient Full-Finetuning of Generative Control Policies [53.42266064673132]
  Generative control policies (GCPs) have emerged as effective parameterizations for robot learning.<n>This work introduces Off-policy Generative Policy Optimization (OGPO), a sample-efficient for finetuning GCPs.<n>OGPO achieves state-of-the-art performance on manipulation tasks spanning multi-task settings, high-precision insertion, and dexterous control.
  arXiv  Detail & Related papers  (2026-05-04T18:36:40Z)
• Diffusion Controller: Framework, Algorithms and Parameterization [54.82539154511621]
  We present a unified control-theoretic view that casts reverse diffusion sampling as state-only control within (generalized) linearly-solvable Markov Decision Processes.<n>Under this framework, control acts by reweighting the pretrained reverse-time transition kernels, balancing terminal objectives against an $f$-divergence cost.<n>Experiments on Stable Diffusion v1.4 show consistent gains in preference-alignment win rates and improved quality-efficiency trade-offs.
  arXiv  Detail & Related papers  (2026-03-07T01:49:59Z)
• Coverage Improvement and Fast Convergence of On-policy Preference Learning [67.36750525893514]
  Online on-policy preference learning algorithms for language model alignment can significantly outperform their offline counterparts.<n>We analyze how the sampling policy's coverage evolves throughout on-policy training.<n>We develop principled on-policy schemes for reward distillation in the general function class setting.
  arXiv  Detail & Related papers  (2026-01-13T10:46:06Z)
• GRPO-Guard: Mitigating Implicit Over-Optimization in Flow Matching via Regulated Clipping [63.33669214116784]
  GRPO-Guard is a simple yet effective enhancement to existing GRPO frameworks.<n>It restores a balanced and step-consistent importance ratio, ensuring that PPO clipping properly constrains harmful updates.<n>It substantially mitigates implicit over-optimization without relying on heavy KL regularization.
  arXiv  Detail & Related papers  (2025-10-25T14:51:17Z)
• Stabilizing Policy Gradients for Sample-Efficient Reinforcement Learning in LLM Reasoning [77.92320830700797]
  Reinforcement Learning has played a central role in enabling reasoning capabilities of Large Language Models.<n>We propose a tractable computational framework that tracks and leverages curvature information during policy updates.<n>The algorithm, Curvature-Aware Policy Optimization (CAPO), identifies samples that contribute to unstable updates and masks them out.
  arXiv  Detail & Related papers  (2025-10-01T12:29:32Z)
• Relative Entropy Pathwise Policy Optimization [66.03329137921949]
  We present an on-policy algorithm that trains Q-value models purely from on-policy trajectories.<n>We show how to combine policies for exploration with constrained updates for stable training, and evaluate important architectural components that stabilize value function learning.
  arXiv  Detail & Related papers  (2025-07-15T06:24:07Z)
• Learning Optimal Deterministic Policies with Stochastic Policy Gradients [62.81324245896716]
  Policy gradient (PG) methods are successful approaches to deal with continuous reinforcement learning (RL) problems. In common practice, convergence (hyper)policies are learned only to deploy their deterministic version. We show how to tune the exploration level used for learning to optimize the trade-off between the sample complexity and the performance of the deployed deterministic policy.
  arXiv  Detail & Related papers  (2024-05-03T16:45:15Z)
• Time-Efficient Reinforcement Learning with Stochastic Stateful Policies [20.545058017790428]
  We present a novel approach for training stateful policies by decomposing the latter into a gradient internal state kernel and a stateless policy. We introduce different versions of the stateful policy gradient theorem, enabling us to easily instantiate stateful variants of popular reinforcement learning algorithms.
  arXiv  Detail & Related papers  (2023-11-07T15:48:07Z)
• Optimization Landscape of Policy Gradient Methods for Discrete-time Static Output Feedback [22.21598324895312]
  This paper analyzes the optimization landscape inherent to policy gradient methods when applied to static output feedback control. We derive novel findings regarding convergence (and nearly dimension-free rate) to stationary points for three policy gradient methods. We provide proof that the vanilla policy gradient method exhibits linear convergence towards local minima when near such minima.
  arXiv  Detail & Related papers  (2023-10-29T14:25:57Z)
• Iteratively Refined Behavior Regularization for Offline Reinforcement Learning [57.10922880400715]
  In this paper, we propose a new algorithm that substantially enhances behavior-regularization based on conservative policy iteration. By iteratively refining the reference policy used for behavior regularization, conservative policy update guarantees gradually improvement. Experimental results on the D4RL benchmark indicate that our method outperforms previous state-of-the-art baselines in most tasks.
  arXiv  Detail & Related papers  (2023-06-09T07:46:24Z)
• A Temporal-Difference Approach to Policy Gradient Estimation [27.749993205038148]
  We propose a new approach of reconstructing the policy gradient from the start state without requiring a particular sampling strategy. By using temporal-difference updates of the gradient critic from an off-policy data stream, we develop the first estimator that sidesteps the distribution shift issue in a model-free way.
  arXiv  Detail & Related papers  (2022-02-04T21:23:33Z)
• Ensuring Monotonic Policy Improvement in Entropy-regularized Value-based Reinforcement Learning [14.325835899564664]
  entropy-regularized value-based reinforcement learning method can ensure the monotonic improvement of policies at each policy update. We propose a novel reinforcement learning algorithm that exploits this lower-bound as a criterion for adjusting the degree of a policy update for alleviating policy oscillation.
  arXiv  Detail & Related papers  (2020-08-25T04:09:18Z)
• DDPG++: Striving for Simplicity in Continuous-control Off-Policy Reinforcement Learning [95.60782037764928]
  We show that simple Deterministic Policy Gradient works remarkably well as long as the overestimation bias is controlled. Second, we pinpoint training instabilities, typical of off-policy algorithms, to the greedy policy update step. Third, we show that ideas in the propensity estimation literature can be used to importance-sample transitions from replay buffer and update policy to prevent deterioration of performance.
  arXiv  Detail & Related papers  (2020-06-26T20:21:12Z)
• Learning Constrained Adaptive Differentiable Predictive Control Policies With Guarantees [1.1086440815804224]
  We present differentiable predictive control (DPC), a method for learning constrained neural control policies for linear systems. We employ automatic differentiation to obtain direct policy gradients by backpropagating the model predictive control (MPC) loss function and constraints penalties through a differentiable closed-loop system dynamics model.
  arXiv  Detail & Related papers  (2020-04-23T14:24:44Z)

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.