MAD: A Magnitude And Direction Policy Parametrization for Stability Constrained Reinforcement Learning

• URL: http://arxiv.org/abs/2504.02565v1
• Date: Thu, 03 Apr 2025 13:26:26 GMT
• Title: MAD: A Magnitude And Direction Policy Parametrization for Stability Constrained Reinforcement Learning
• Authors: Luca Furieri, Sucheth Shenoy, Danilo Saccani, Andrea Martin, Giancarlo Ferrari Trecate,
• Abstract summary: We introduce magnitude and direction (MAD) policies, a policy parameterization for reinforcement learning (RL)<n>MAD policies introduce explicit feedback on state-dependent features without compromising closed-loop stability.<n>We show that MAD policies trained with deep deterministic policy gradient (DDPG) methods generalize to unseen scenarios.
• Score: 1.712670816823812
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We introduce magnitude and direction (MAD) policies, a policy parameterization for reinforcement learning (RL) that preserves Lp closed-loop stability for nonlinear dynamical systems. Although complete in their ability to describe all stabilizing controllers, methods based on nonlinear Youla and system-level synthesis are significantly affected by the difficulty of parameterizing Lp-stable operators. In contrast, MAD policies introduce explicit feedback on state-dependent features - a key element behind the success of RL pipelines - without compromising closed-loop stability. This is achieved by describing the magnitude of the control input with a disturbance-feedback Lp-stable operator, while selecting its direction based on state-dependent features through a universal function approximator. We further characterize the robust stability properties of MAD policies under model mismatch. Unlike existing disturbance-feedback policy parameterizations, MAD policies introduce state-feedback components compatible with model-free RL pipelines, ensuring closed-loop stability without requiring model information beyond open-loop stability. Numerical experiments show that MAD policies trained with deep deterministic policy gradient (DDPG) methods generalize to unseen scenarios, matching the performance of standard neural network policies while guaranteeing closed-loop stability by design.

Related papers

• Relative Entropy Pathwise Policy Optimization [56.86405621176669]
  We show how to construct a value-gradient driven, on-policy algorithm that allow training Q-value models purely from on-policy data.<n>We propose Relative Entropy Pathwise Policy Optimization (REPPO), an efficient on-policy algorithm that combines the sample-efficiency of pathwise policy gradients with the simplicity and minimal memory footprint of standard on-policy learning.
  arXiv  Detail & Related papers  (2025-07-15T06:24:07Z)
• Learning Stabilizing Policies via an Unstable Subspace Representation [0.0]
  We study the problem of learning to stabilize (LTS) a linear time-invariant (LTI) system.<n>We propose a two-phase approach that first learns the left unstable subspace of the system.<n>We demonstrate that operating on the unstable subspace reduces sample complexity.
  arXiv  Detail & Related papers  (2025-05-02T15:34:36Z)
• Learning to Boost the Performance of Stable Nonlinear Systems [0.0]
  We tackle the performance-boosting problem with closed-loop stability guarantees. Our methods enable learning over arbitrarily deep neural network classes of performance-boosting controllers for stable nonlinear systems.
  arXiv  Detail & Related papers  (2024-05-01T21:11:29Z)
• Synthesizing Stable Reduced-Order Visuomotor Policies for Nonlinear Systems via Sums-of-Squares Optimization [28.627377507894003]
  We present a method for noise-feedback, reduced-order output-of-control-perception policies for control observations of nonlinear systems. We show that when these systems from images can fail to reliably stabilize, our approach can provide stability guarantees.
  arXiv  Detail & Related papers  (2023-04-24T19:34:09Z)
• Learning Over Contracting and Lipschitz Closed-Loops for Partially-Observed Nonlinear Systems (Extended Version) [1.2430809884830318]
  This paper presents a policy parameterization for learning-based control on nonlinear, partially-observed dynamical systems. We prove that the resulting Youla-REN parameterization automatically satisfies stability (contraction) and user-tunable robustness (Lipschitz) conditions. We find that the Youla-REN performs similarly to existing learning-based and optimal control methods while also ensuring stability and exhibiting improved robustness to adversarial disturbances.
  arXiv  Detail & Related papers  (2023-04-12T23:55:56Z)
• Hallucinated Adversarial Control for Conservative Offline Policy Evaluation [64.94009515033984]
  We study the problem of conservative off-policy evaluation (COPE) where given an offline dataset of environment interactions, we seek to obtain a (tight) lower bound on a policy's performance. We introduce HAMBO, which builds on an uncertainty-aware learned model of the transition dynamics. We prove that the resulting COPE estimates are valid lower bounds, and, under regularity conditions, show their convergence to the true expected return.
  arXiv  Detail & Related papers  (2023-03-02T08:57:35Z)
• Bounded Robustness in Reinforcement Learning via Lexicographic Objectives [54.00072722686121]
  Policy robustness in Reinforcement Learning may not be desirable at any cost. We study how policies can be maximally robust to arbitrary observational noise. We propose a robustness-inducing scheme, applicable to any policy algorithm, that trades off expected policy utility for robustness.
  arXiv  Detail & Related papers  (2022-09-30T08:53:18Z)
• KCRL: Krasovskii-Constrained Reinforcement Learning with Guaranteed Stability in Nonlinear Dynamical Systems [66.9461097311667]
  We propose a model-based reinforcement learning framework with formal stability guarantees. The proposed method learns the system dynamics up to a confidence interval using feature representation. We show that KCRL is guaranteed to learn a stabilizing policy in a finite number of interactions with the underlying unknown system.
  arXiv  Detail & Related papers  (2022-06-03T17:27:04Z)
• Neural System Level Synthesis: Learning over All Stabilizing Policies for Nonlinear Systems [0.0]
  We propose a Neural SLS (Neur-SLS) approach guaranteeing closed-loop stability during and after parameter optimization. We exploit recent Deep Neural Network (DNN) models based on Recurrent Equilibrium Networks (RENs) to learn over a rich class of nonlinear stable operators.
  arXiv  Detail & Related papers  (2022-03-22T15:22:31Z)
• Supported Policy Optimization for Offline Reinforcement Learning [74.1011309005488]
  Policy constraint methods to offline reinforcement learning (RL) typically utilize parameterization or regularization. Regularization methods reduce the divergence between the learned policy and the behavior policy. This paper presents Supported Policy OpTimization (SPOT), which is directly derived from the theoretical formalization of the density-based support constraint.
  arXiv  Detail & Related papers  (2022-02-13T07:38:36Z)
• Youla-REN: Learning Nonlinear Feedback Policies with Robust Stability Guarantees [5.71097144710995]
  This paper presents a parameterization of nonlinear controllers for uncertain systems building on a recently developed neural network architecture. The proposed framework has "built-in" guarantees of stability, i.e., all policies in the search space result in a contracting (globally exponentially stable) closed-loop system.
  arXiv  Detail & Related papers  (2021-12-02T13:52:37Z)
• On the Sample Complexity and Metastability of Heavy-tailed Policy Search in Continuous Control [47.71156648737803]
  Reinforcement learning is a framework for interactive decision-making with incentives sequentially revealed across time without a system dynamics model. We characterize a defined defined chain, identifying that policies associated with Levy Processes of a tail index yield to wider peaks.
  arXiv  Detail & Related papers  (2021-06-15T20:12:44Z)
• Improper Learning with Gradient-based Policy Optimization [62.50997487685586]
  We consider an improper reinforcement learning setting where the learner is given M base controllers for an unknown Markov Decision Process. We propose a gradient-based approach that operates over a class of improper mixtures of the controllers.
  arXiv  Detail & Related papers  (2021-02-16T14:53:55Z)

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.