This paper addresses the challenge of solving Constrained Markov Decision Processes (CMDPs) with $d > 1$ constraints when the transition dynamics are unknown, but samples can be drawn from a generative model. We propose a model-based algorithm for infinite horizon CMDPs with multiple constraints in the tabular setting, aiming to derive and prove sample complexity bounds for learning near-optimal policies. Our approach tackles both the relaxed and strict feasibility settings, where relaxed feasibility allows some constraint violations, and strict feasibility requires adherence to all constraints. The main contributions include the development of the algorithm and the derivation of sample complexity bounds for both settings. For the relaxed feasibility setting we show that our algorithm requires $\tilde{\mathcal{O}} \left( \frac{d |\mathcal{S}| |\mathcal{A}| \log(1/δ)}{(1-γ)^3ε^2} \right)$ samples to return $ε$-optimal policy, while in the strict feasibility setting it requires $\tilde{\mathcal{O}} \left( \frac{d^3 |\mathcal{S}| |\mathcal{A}| \log(1/δ)}{(1-γ)^5ε^2{ζ_{\mathbf{c}}^*}^2} \right)$ samples.
A Policy Gradient Primal-Dual Algorithm for Constrained MDPs with Uniform PAC Guarantees
The University of Tokyo（东京大学）；OMRON SINIC X（欧姆龙SINIC X公司）；Mizuho–DL Financial Technology（瑞穗-DL金融科技公司）；Nara Institute of Science and Technology（奈良先端科学技术大学院大学）；RIKEN Center for Advanced Intelligence Project（理化学研究所革新智能综合研究中心）；Osaka University（大阪大学）；Kyoto University（京都大学）
Purdue University（普渡大学）；US Army Research Laboratory（美国陆军研究实验室）
Reinforcement learning is widely used in applications where one needs to perform sequential decisions while interacting with the environment. The problem becomes more challenging when the decision requirement includes satisfying some safety constraints. The problem is mathematically formulated as constrained Markov decision process (CMDP). In the literature, various algorithms are available to solve CMDP problems in a model-free manner to achieve $ε$-optimal cumulative reward with $ε$ feasible policies. An $ε$-feasible policy implies that it suffers from constraint violation. An important ques…
Finite-Time Complexity of Online Primal-Dual Natural Actor-Critic Algorithm for Constrained Markov Decision Processes
School of Electrical and Computer Engineering, Georgia Institute of Technology（佐治亚理工学院电气与计算机工程学院）；Bradley Department of Electrical and Computer Engineering, Virginia Tech（弗吉尼亚理工大学布拉德利电气与计算机工程学院）
We study the problem of learning policy of an infinite-horizon, discounted cost, Markov decision process (MDP) with a large number of states. We compute the actions of a policy that is nearly as good as a policy chosen by a suitable oracle from a given mixture policy class characterized by the convex hull of a set of known base policies. To learn the coefficients of the mixture model, we recast the problem as an approximate linear programming (ALP) formulation for MDPs, where the feature vectors correspond to the occupation measures of the base policies defined on the state-action space. We th…
Convergence and sample complexity of natural policy gradient primal-dual methods for constrained MDPs
University of Tennessee, Knoxville（田纳西大学）；University of Maryland, College Park（马里兰大学）；Apple Inc.（苹果公司）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of Southern California（南加州大学）
Constrained Markov Decision Process (CMDP) is a natural framework for reinforcement learning tasks with safety constraints, where agents learn a policy that maximizes the long-term reward while satisfying the constraints on the long-term cost. A canonical approach for solving CMDPs is the primal-dual method which updates parameters in primal and dual spaces in turn. Existing methods for CMDPs only use on-policy data for dual updates, which results in sample inefficiency and slow convergence. In this paper, we propose a policy search method for CMDPs called Accelerated Primal-Dual Optimization …
We propose a new stochastic primal-dual optimization algorithm for planning in a large discounted Markov decision process with a generative model and linear function approximation. Assuming that the feature map approximately satisfies standard realizability and Bellman-closedness conditions and also that the feature vectors of all state-action pairs are representable as convex combinations of a small core set of state-action pairs, we show that our method outputs a near-optimal policy after a polynomial number of queries to the generative model. Our method is computationally efficient and come…
Georgia Institute of Technology（佐治亚理工学院）；The Ohio State University（俄亥俄州立大学）；University at Buffalo, The State University of New York（纽约州立大学布法罗分校）
The problem of constrained Markov decision process (CMDP) is investigated, where an agent aims to maximize the expected accumulated discounted reward subject to multiple constraints on its utilities/costs. A new primal-dual approach is proposed with a novel integration of three ingredients: entropy regularized policy optimizer, dual variable regularizer, and Nesterov's accelerated gradient descent dual optimizer, all of which are critical to achieve a faster convergence. The finite-time error bound of the proposed approach is characterized. Despite the challenge of the nonconcave objective sub…
We address the problem of finding the optimal policy of a constrained Markov decision process (CMDP) using a gradient descent-based algorithm. Previous results have shown that a primal-dual approach can achieve an $\mathcal{O}(1/\sqrt{T})$ global convergence rate for both the optimality gap and the constraint violation. We propose a new algorithm called policy mirror descent-primal dual (PMD-PD) algorithm that can provably achieve a faster $\mathcal{O}(\log(T)/T)$ convergence rate for both the optimality gap and the constraint violation. For the primal (policy) update, the PMD-PD algorithm uti…
University of Michigan（密歇根大学）；University of Southern California（南加州大学）；Princeton University（普林斯顿大学）；Northwestern University（西北大学）
We consider online learning for episodic stochastically constrained Markov decision processes (CMDPs), which plays a central role in ensuring the safety of reinforcement learning. Here the loss function can vary arbitrarily across the episodes, and both the loss received and the budget consumption are revealed at the end of each episode. Previous works solve this problem under the restrictive assumption that the transition model of the Markov decision processes (MDPs) is known a priori and establish regret bounds that depend polynomially on the cardinalities of the state space $\mathcal{S}$ an…
The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；JD Explore Academy（京东探索研究院）
We study the convergence of the actor-critic algorithm with nonlinear function approximation under a nonconvex-nonconcave primal-dual formulation. Stochastic gradient descent ascent is applied with an adaptive proximal term for robust learning rates. We show the first efficient convergence result with primal-dual actor-critic with a convergence rate of $\mathcal{O}\left(\sqrt{\frac{\ln \left(N d G^2 \right)}{N}}\right)$ under Markovian sampling, where $G$ is the element-wise maximum of the gradient, $N$ is the number of iterations, and $d$ is the dimension of the gradient. Our result is presen…
We consider large-scale Markov decision processes with an unknown cost function and address the problem of learning a policy from a finite set of expert demonstrations. We assume that the learner is not allowed to interact with the expert and has no access to reinforcement signal of any kind. Existing inverse reinforcement learning methods come with strong theoretical guarantees, but are computationally expensive, while state-of-the-art policy optimization algorithms achieve significant empirical success, but are hampered by limited theoretical understanding. To bridge the gap between theory a…
This paper studies how to train machine-learning models that directly approximate the optimal solutions of constrained optimization problems. This is an empirical risk minimization under constraints, which is challenging as training must balance optimality and feasibility conditions. Supervised learning methods often approach this challenge by training the model on a large collection of pre-solved instances. This paper takes a different route and proposes the idea of Primal-Dual Learning (PDL), a self-supervised training method that does not require a set of pre-solved instances or an optimiza…
Consider the problem of approximating the optimal policy of a Markov decision process (MDP) by sampling state transitions. In contrast to existing reinforcement learning methods that are based on successive approximations to the nonlinear Bellman equation, we propose a Primal-Dual $π$ Learning method in light of the linear duality between the value and policy. The $π$ learning method is model-free and makes primal-dual updates to the policy and value vectors as new data are revealed. For infinite-horizon undiscounted Markov decision process with finite state space $S$ and finite action space $…
Risk-aware control, though with promise to tackle unexpected events, requires a known exact dynamical model. In this work, we propose a model-free framework to learn a risk-aware controller with a focus on the linear system. We formulate it as a discrete-time infinite-horizon LQR problem with a state predictive variance constraint. To solve it, we parameterize the policy with a feedback gain pair and leverage primal-dual methods to optimize it by solely using data. We first study the optimization landscape of the Lagrangian function and establish the strong duality in spite of its non-convex n…
School of Information Science and Technology, ShanghaiTech University（上海科技大学信息科学与技术学院）；Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences（中国科学院上海微系统与信息技术研究所）；University of Washington（华盛顿大学）；Tongji University（同济大学）；University of Chinese Academy of Sciences（中国科学院大学）
In this paper, we propose a novel primal-dual inexact gradient projection method for nonlinear optimization problems with convex-set constraint. This method only needs inexact computation of the projections onto the convex set for each iteration, consequently reducing the computational cost for projections per iteration. This feature is attractive especially for solving problems where the projections are computationally not easy to calculate. Global convergence guarantee and O(1/k) ergodic convergence rate of the optimality residual are provided under loose assumptions. We apply our proposed s…
Cornell University（康奈尔大学）；California Institute of Technology（加州理工学院）
This paper develops a new storage-optimal algorithm that provably solves generic semidefinite programs (SDPs) in standard form. This method is particularly effective for weakly constrained SDPs. The key idea is to formulate an approximate complementarity principle: Given an approximate solution to the dual SDP, the primal SDP has an approximate solution whose range is contained in the eigenspace with small eigenvalues of the dual slack matrix. For weakly constrained SDPs, this eigenspace has very low dimension, so this observation significantly reduces the search space for the primal solution.…
We introduce a new sequential subspace optimization method for large-scale saddle-point problems. It solves iteratively a sequence of auxiliary saddle-point problems in low-dimensional subspaces, spanned by directions derived from first-order information over the primal \emph{and} dual variables. Proximal regularization is further deployed to stabilize the optimization process. Experimental results demonstrate significantly better convergence relative to popular first-order methods. We analyze the influence of the subspace on the convergence of the algorithm, and assess its performance in vari…
State-of-the-art deep neural networks are sensitive to small input perturbations. Since the discovery of this intriguing vulnerability, many defence methods have been proposed that attempt to improve robustness to adversarial noise. Fast and accurate attacks are required to compare various defence methods. However, evaluating adversarial robustness has proven to be extremely challenging. Existing norm minimisation adversarial attacks require thousands of iterations (e.g. Carlini & Wagner attack), are limited to the specific norms (e.g. Fast Adaptive Boundary), or produce sub-optimal results (e…
