Robust Lagrangian and Adversarial Policy Gradient for Robust Constrained Markov Decision Processes

• URL: http://arxiv.org/abs/2308.11267v3
• Date: Wed, 15 May 2024 14:46:34 GMT
• Title: Robust Lagrangian and Adversarial Policy Gradient for Robust Constrained Markov Decision Processes
• Authors: David M. Bossens,
• Abstract summary: This paper introduces two algorithms, called RCPG with Robust Lagrangian and Adversarial RCPG. RCPG with Robust Lagrangian modifies RCPG by taking the worst-case dynamics based on the Lagrangian rather than either the value or the constraint. Adversarial RCPG also formulates the worst-case dynamics based on the Lagrangian but learns this directly and incrementally as an adversarial policy.
• Score: 5.167069404528051
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The robust constrained Markov decision process (RCMDP) is a recent task-modelling framework for reinforcement learning that incorporates behavioural constraints and that provides robustness to errors in the transition dynamics model through the use of an uncertainty set. Simulating RCMDPs requires computing the worst-case dynamics based on value estimates for each state, an approach which has previously been used in the Robust Constrained Policy Gradient (RCPG). Highlighting potential downsides of RCPG such as not robustifying the full constrained objective and the lack of incremental learning, this paper introduces two algorithms, called RCPG with Robust Lagrangian and Adversarial RCPG. RCPG with Robust Lagrangian modifies RCPG by taking the worst-case dynamics based on the Lagrangian rather than either the value or the constraint. Adversarial RCPG also formulates the worst-case dynamics based on the Lagrangian but learns this directly and incrementally as an adversarial policy through gradient descent rather than indirectly and abruptly through constrained optimisation on a sorted value list. A theoretical analysis first derives the Lagrangian policy gradient for the policy optimisation of both proposed algorithms and then the adversarial policy gradient to learn the adversary for Adversarial RCPG. Empirical experiments injecting perturbations in inventory management and safe navigation tasks demonstrate the competitive performance of both algorithms compared to traditional RCPG variants as well as non-robust and non-constrained ablations. In particular, Adversarial RCPG ranks among the top two performing algorithms on all tests.

Related papers

• Deep Reinforcement Learning Algorithms for Option Hedging [0.20482269513546458]
  We compare the performance of eight Deep Reinforcement Learning (DRL) algorithms in the context of dynamic hedging. MCPG is the only algorithm to outperform the Black-Scholes delta hedge baseline with the allotted computational budget.
  arXiv  Detail & Related papers  (2025-04-07T21:32:14Z)
• Robust Deterministic Policy Gradient for Disturbance Attenuation and Its Application to Quadrotor Control [5.084000938840218]
  This paper proposes a reinforcement learning algorithm called Robust Deterministic Policy Gradient (RDPG) RDPG formulates the $H_infty$ control problem as a two-player zero-sum dynamic game. We then employ deterministic policy gradient (DPG) and its deep reinforcement learning counterpart to train a robust control policy with effective disturbance attenuation.
  arXiv  Detail & Related papers  (2025-02-28T13:58:22Z)
• Deterministic Policy Gradient Primal-Dual Methods for Continuous-Space Constrained MDPs [82.34567890576423]
  We develop a deterministic policy gradient primal-dual method to find an optimal deterministic policy with non-asymptotic convergence. We prove that the primal-dual iterates of D-PGPD converge at a sub-linear rate to an optimal regularized primal-dual pair. To the best of our knowledge, this appears to be the first work that proposes a deterministic policy search method for continuous-space constrained MDPs.
  arXiv  Detail & Related papers  (2024-08-19T14:11:04Z)
• 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)
• Global Convergence of Receding-Horizon Policy Search in Learning Estimator Designs [3.0811185425377743]
  We introduce the receding-horizon policy estimator (RHPG) algorithm. RHPG is the first algorithm with provable global convergence in learning optimal linear policy estimator.
  arXiv  Detail & Related papers  (2023-09-09T16:03:49Z)
• Multi-Objective Policy Gradients with Topological Constraints [108.10241442630289]
  We present a new algorithm for a policy gradient in TMDPs by a simple extension of the proximal policy optimization (PPO) algorithm. We demonstrate this on a real-world multiple-objective navigation problem with an arbitrary ordering of objectives both in simulation and on a real robot.
  arXiv  Detail & Related papers  (2022-09-15T07:22:58Z)
• 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)
• Causal Policy Gradients [6.123324869194195]
  Causal policy gradients (CPGs) provide a common framework for analysing key state-of-the-art algorithms. CPGs are shown to generalise traditional policy gradients, and yield a principled way of incorporating prior knowledge of a problem domain's generative processes.
  arXiv  Detail & Related papers  (2021-02-20T14:51:12Z)
• Risk-Sensitive Deep RL: Variance-Constrained Actor-Critic Provably Finds Globally Optimal Policy [95.98698822755227]
  We make the first attempt to study risk-sensitive deep reinforcement learning under the average reward setting with the variance risk criteria. We propose an actor-critic algorithm that iteratively and efficiently updates the policy, the Lagrange multiplier, and the Fenchel dual variable.
  arXiv  Detail & Related papers  (2020-12-28T05:02:26Z)
• Zeroth-order Deterministic Policy Gradient [116.87117204825105]
  We introduce Zeroth-order Deterministic Policy Gradient (ZDPG) ZDPG approximates policy-reward gradients via two-point evaluations of the $Q$function. New finite sample complexity bounds for ZDPG improve upon existing results by up to two orders of magnitude.
  arXiv  Detail & Related papers  (2020-06-12T16:52:29Z)
• Discrete Action On-Policy Learning with Action-Value Critic [72.20609919995086]
  Reinforcement learning (RL) in discrete action space is ubiquitous in real-world applications, but its complexity grows exponentially with the action-space dimension. We construct a critic to estimate action-value functions, apply it on correlated actions, and combine these critic estimated action values to control the variance of gradient estimation. These efforts result in a new discrete action on-policy RL algorithm that empirically outperforms related on-policy algorithms relying on variance control techniques.
  arXiv  Detail & Related papers  (2020-02-10T04:23:09Z)

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.