ISOPO: Proximal policy gradients without pi-old

• URL: http://arxiv.org/abs/2512.23353v2
• Date: Tue, 30 Dec 2025 03:46:06 GMT
• Title: ISOPO: Proximal policy gradients without pi-old
• Authors: Nilin Abrahamsen,
• Abstract summary: ISOPO is an efficient method to approximate the natural policy gradient in a single step.<n>It can be implemented with negligible computational overhead compared to vanilla REINFORCE.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This note introduces Isometric Policy Optimization (ISOPO), an efficient method to approximate the natural policy gradient in a single gradient step. In comparison, existing proximal policy methods such as GRPO or CISPO use multiple gradient steps with variants of importance ratio clipping to approximate a natural gradient step relative to a reference policy. In its simplest form, ISOPO normalizes the log-probability gradient of each sequence in the Fisher metric before contracting with the advantages. Another variant of ISOPO transforms the microbatch advantages based on the neural tangent kernel in each layer. ISOPO applies this transformation layer-wise in a single backward pass and can be implemented with negligible computational overhead compared to vanilla REINFORCE.

Related papers

• Rank-1 Approximation of Inverse Fisher for Natural Policy Gradients in Deep Reinforcement Learning [17.531852538779372]
  We show that a rank-1 approximation to inverse-FIM converges faster than policy gradients.<n>We benchmark our method on a diverse set of environments and show that it achieves superior performance to standard actor-critic and trust-region baselines.
  arXiv  Detail & Related papers  (2026-01-26T16:02:18Z)
• Reinforcement Learning in POMDP's via Direct Gradient Ascent [21.715823431124235]
  We introduce GPOMDP, a REINFORCE-like algorithm for estimating an approximation to the gradient of the average reward.<n>We show how GPOMDP can be used in a conjugate-gradient procedure to find local optima of the average reward.
  arXiv  Detail & Related papers  (2025-12-02T03:50:06Z)
• Generalized Gradient Norm Clipping & Non-Euclidean $(L_0,L_1)$-Smoothness [51.302674884611335]
  This work introduces a hybrid non-Euclidean optimization method which generalizes norm clipping by combining steepest descent and conditional gradient approaches.<n>We discuss how to instantiate the algorithms for deep learning and demonstrate their properties on image classification and language modeling.
  arXiv  Detail & Related papers  (2025-06-02T17:34:29Z)
• Policy Gradient with Active Importance Sampling [55.112959067035916]
  Policy gradient (PG) methods significantly benefit from IS, enabling the effective reuse of previously collected samples. However, IS is employed in RL as a passive tool for re-weighting historical samples. We look for the best behavioral policy from which to collect samples to reduce the policy gradient variance.
  arXiv  Detail & Related papers  (2024-05-09T09:08:09Z)
• Clipped-Objective Policy Gradients for Pessimistic Policy Optimization [3.2996723916635275]
  Policy gradient methods seek to produce monotonic improvement through bounded changes in policy outputs. In this work, we find that the performance of PPO, when applied to continuous action spaces, may be consistently improved through a simple change in objective. We show that the clipped-objective policy gradient (COPG) objective is on average "pessimistic" compared to both the PPO objective and (2) this pessimism promotes enhanced exploration.
  arXiv  Detail & Related papers  (2023-11-10T03:02:49Z)
• 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)
• Last-Iterate Convergent Policy Gradient Primal-Dual Methods for Constrained MDPs [107.28031292946774]
  We study the problem of computing an optimal policy of an infinite-horizon discounted Markov decision process (constrained MDP) We develop two single-time-scale policy-based primal-dual algorithms with non-asymptotic convergence of their policy iterates to an optimal constrained policy. To the best of our knowledge, this work appears to be the first non-asymptotic policy last-iterate convergence result for single-time-scale algorithms in constrained MDPs.
  arXiv  Detail & Related papers  (2023-06-20T17:27:31Z)
• Linear Convergence of Natural Policy Gradient Methods with Log-Linear Policies [115.86431674214282]
  We consider infinite-horizon discounted Markov decision processes and study the convergence rates of the natural policy gradient (NPG) and the Q-NPG methods with the log-linear policy class. We show that both methods attain linear convergence rates and $mathcalO (1/epsilon2)$ sample complexities using a simple, non-adaptive geometrically increasing step size.
  arXiv  Detail & Related papers  (2022-10-04T06:17:52Z)
• Anchor-Changing Regularized Natural Policy Gradient for Multi-Objective Reinforcement Learning [17.916366827429034]
  We study policy optimization for Markov decision processes (MDPs) with multiple reward value functions. We propose an Anchor-changing Regularized Natural Policy Gradient framework, which can incorporate ideas from well-performing first-order methods.
  arXiv  Detail & Related papers  (2022-06-10T21:09:44Z)
• Bregman Gradient Policy Optimization [97.73041344738117]
  We design a Bregman gradient policy optimization for reinforcement learning based on Bregman divergences and momentum techniques. VR-BGPO reaches the best complexity $tilde(epsilon-3)$ for finding an $epsilon$stationary point only requiring one trajectory at each iteration.
  arXiv  Detail & Related papers  (2021-06-23T01:08:54Z)
• On the Linear convergence of Natural Policy Gradient Algorithm [5.027714423258537]
  Recent interest in Reinforcement Learning has motivated the study of methods inspired by optimization. Among these is the Natural Policy Gradient, which is a mirror descent variant for MDPs. We present improved finite time convergence bounds, and show that this algorithm has geometric convergence rate.
  arXiv  Detail & Related papers  (2021-05-04T11:26:12Z)
• Softmax Policy Gradient Methods Can Take Exponential Time to Converge [60.98700344526674]
  The softmax policy gradient (PG) method is arguably one of the de facto implementations of policy optimization in modern reinforcement learning. We demonstrate that softmax PG methods can take exponential time -- in terms of $mathcalS|$ and $frac11-gamma$ -- to converge.
  arXiv  Detail & Related papers  (2021-02-22T18:56:26Z)

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.