We consider the problem of finding optimal policies for a Markov Decision Process with almost sure constraints on state transitions and action triplets. We define value and action-value functions that satisfy a barrier-based decomposition which allows for the identification of feasible policies independently of the reward process. We prove that, given a policy π, certifying whether certain state-action pairs lead to feasible trajectories under π is equivalent to solving an auxiliary problem aimed at finding the probability of performing an unfeasible transition. Using this interpretation,we develop a Barrier-learning algorithm, based on Q-Learning, that identifies such unsafe state-action pairs. Our analysis motivates the need to enhance the Reinforcement Learning (RL) framework with an additional signal, besides rewards, called here damage function that provides feasibility information and enables the solution of RL problems with model-free constraints. Moreover, our Barrier-learning algorithm wraps around existing RL algorithms, such as Q-Learning and SARSA, giving them the ability to solve almost-surely constrained problems.
Michigan State University（密歇根州立大学）；Ford Research and Innovation Center, Ford Motor Company（福特汽车公司研究与创新中心）
Reinforcement learning (RL) agents with pre-specified reward functions cannot provide guaranteed safety across variety of circumstances that an uncertain system might encounter. To guarantee performance while assuring satisfaction of safety constraints across variety of circumstances, an assured autonomous control framework is presented in this paper by empowering RL algorithms with metacognitive learning capabilities. More specifically, adapting the reward function parameters of the RL agent is performed in a metacognitive decision-making layer to assure the feasibility of RL agent. That is, …
Johns Hopkins University（约翰斯·霍普金斯大学）；Universidad de la República（共和国大学）
In this work we address the problem of finding feasible policies for Constrained Markov Decision Processes under probability one constraints. We argue that stationary policies are not sufficient for solving this problem, and that a rich class of policies can be found by endowing the controller with a scalar quantity, so called budget, that tracks how close the agent is to violating the constraint. We show that the minimal budget required to act safely can be obtained as the smallest fixed point of a Bellman-like operator, for which we analyze its convergence properties. We also show how to lea…
Huawei R&D UK（华为英国研发部）；Technische Universität Darmstadt（达姆施塔特工业大学）；University College London（伦敦大学学院）
Satisfying safety constraints almost surely (or with probability one) can be critical for the deployment of Reinforcement Learning (RL) in real-life applications. For example, plane landing and take-off should ideally occur with probability one. We address the problem by introducing Safety Augmented (Saute) Markov Decision Processes (MDPs), where the safety constraints are eliminated by augmenting them into the state-space and reshaping the objective. We show that Saute MDP satisfies the Bellman equation and moves us closer to solving Safe RL with constraints satisfied almost surely. We argue …
In constrained reinforcement learning (C-RL), an agent seeks to learn from the environment a policy that maximizes the expected cumulative reward while satisfying minimum requirements in secondary cumulative reward constraints. Several algorithms rooted in sampled-based primal-dual methods have been recently proposed to solve this problem in policy space. However, such methods are based on stochastic gradient descent ascent algorithms whose trajectories are connected to the optimal policy only after a mixing output stage that depends on the algorithm's history. As a result, there is a mismatch…
University of Liverpool（利物浦大学）；University of Colorado Boulder（科罗拉多大学博尔德分校）
We provide the first solution for model-free reinforcement learning of ω-regular objectives for Markov decision processes (MDPs). We present a constructive reduction from the almost-sure satisfaction of ω-regular objectives to an almost- sure reachability problem and extend this technique to learning how to control an unknown model so that the chance of satisfying the objective is maximized. A key feature of our technique is the compilation of ω-regular properties into limit- deterministic Buechi automata instead of the traditional Rabin automata; this choice sidesteps difficulties that have m…
Watkins' and Dayan's Q-learning is a model-free reinforcement learning algorithm that iteratively refines an estimate for the optimal action-value function of an MDP by stochastically "visiting" many state-ation pairs [Watkins and Dayan, 1992]. Variants of the algorithm lie at the heart of numerous recent state-of-the-art achievements in reinforcement learning, including the superhuman Atari-playing deep Q-network [Mnih et al., 2015]. The goal of this paper is to reproduce a precise and (nearly) self-contained proof that Q-learning converges. Much of the available literature leverages powerful…
Naval Surface Warfare Center（海军水面作战中心）；Naval Sea Systems Command（海军海上系统司令部）；University of Florida（佛罗里达大学）
In this paper, we consider the problem of adjusting the exploration rate when using value-of-information-based exploration. We do this by converting the value-of-information optimization into a problem of finding equilibria of a flow for a changing exploration rate. We then develop an efficient path-following scheme for converging to these equilibria and hence uncovering optimal action-selection policies. Under this scheme, the exploration rate is automatically adapted according to the agent's experiences. Global convergence is theoretically assured. We first evaluate our exploration-rate adap…
Cornell University（康奈尔大学）；U.S. Army Research Laboratory（美国陆军研究实验室）
This paper considers policy search in continuous state-action reinforcement learning problems. Typically, one computes search directions using a classic expression for the policy gradient called the Policy Gradient Theorem, which decomposes the gradient of the value function into two factors: the score function and the Q-function. This paper presents four results:(i) an alternative policy gradient theorem using weak (measure-valued) derivatives instead of score-function is established; (ii) the stochastic gradient estimates thus derived are shown to be unbiased and to yield algorithms that con…
How to best explore in domains with sparse, delayed, and deceptive rewards is an important open problem for reinforcement learning (RL). This paper considers one such domain, the recently-proposed multi-agent benchmark of Pommerman. This domain is very challenging for RL --- past work has shown that model-free RL algorithms fail to achieve significant learning without artificially reducing the environment's complexity. In this paper, we illuminate reasons behind this failure by providing a thorough analysis on the hardness of random exploration in Pommerman. While model-free random exploration…
We propose a fully distributed actor-critic architecture, named Diff-DAC, with application to multitask reinforcement learning (MRL). During the learning process, agents communicate their value and policy parameters to their neighbours, diffusing the information across a network of agents with no need for a central station. Each agent can only access data from its local task, but aims to learn a common policy that performs well for the whole set of tasks. The architecture is scalable, since the computational and communication cost per agent depends on the number of neighbours rather than the o…
MIT Computer Science and Artificial Intelligence Laboratory（麻省理工学院计算机科学与人工智能实验室）
We address the problem of efficient exploration for transition model learning in the relational model-based reinforcement learning setting without extrinsic goals or rewards. Inspired by human curiosity, we propose goal-literal babbling (GLIB), a simple and general method for exploration in such problems. GLIB samples relational conjunctive goals that can be understood as specific, targeted effects that the agent would like to achieve in the world, and plans to achieve these goals using the transition model being learned. We provide theoretical guarantees showing that exploration with GLIB wil…
Reinforcement learning can interact with the environment and is suitable for applications in decision control systems. Therefore, we used the reinforcement learning method to establish a foreign exchange transaction, avoiding the long-standing problem of unstable trends in deep learning predictions. In the system design, we optimized the Sure-Fire statistical arbitrage policy, set three different actions, encoded the continuous price over a period of time into a heat-map view of the Gramian Angular Field (GAF) and compared the Deep Q Learning (DQN) and Proximal Policy Optimization (PPO) algori…
We propose a successive convex approximation based off-policy optimization (SCAOPO) algorithm to solve the general constrained reinforcement learning problem, which is formulated as a constrained Markov decision process (CMDP) in the context of average cost. The SCAOPO is based on solving a sequence of convex objective/feasibility optimization problems obtained by replacing the objective and constraint functions in the original problems with convex surrogate functions. At each iteration, the convex surrogate problem can be efficiently solved by Lagrange dual method even the policy is parameter…
Key Laboratory of System Software (Chinese Academy of Sciences), Institute of Software, Chinese Academy of Sciences, Beijing, China（中国科学院系统软件重点实验室，软件研究所，中国科学院，北京，中国）；University of Chinese Academy of Sciences, Beijing, China（中国科学院大学，北京，中国）；School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences, Beijing, China（中国科学院大学交叉学科学院，中国科学院大学，北京，中国）
Fraunhofer Center for Machine Learning（弗劳恩霍夫机器学习中心）；Fraunhofer IOSB-INA（弗劳恩霍夫IOSB-INA研究所）；University of Groningen（格罗宁根大学）
Reinforcement learning (RL) for traffic signal control (TSC) has shown better performance in simulation for controlling the traffic flow of intersections than conventional approaches. However, due to several challenges, no RL-based TSC has been deployed in the field yet. One major challenge for real-world deployment is to ensure that all safety requirements are met at all times during operation. We present an approach to ensure safety in a real-world intersection by using an action space that is safe by design. The action space encompasses traffic phases, which represent the combination of non…
Institute for Computational & Mathematical Engineering, Stanford（斯坦福大学计算与数学工程研究所）；Department of Electrical Engineering, Stanford（斯坦福大学电气工程系）；DeepMind（深度思维公司）；Google（谷歌公司）
Policy gradient methods are among the most effective methods for large-scale reinforcement learning, and their empirical success has prompted several works that develop the foundation of their global convergence theory. However, prior works have either required exact gradients or state-action visitation measure based mini-batch stochastic gradients with a diverging batch size, which limit their applicability in practical scenarios. In this paper, we consider classical policy gradient methods that compute an approximate gradient with a single trajectory or a fixed size mini-batch of trajectorie…
Temporal-Causal Inference for Reinforcement Learning via Automata Learning
Research Center Trustworthy Data Science and Security, TU Dortmund University（多特蒙德工业大学可信数据科学与安全研究中心）；School for Engineering of Matter, Transport, and Energy at Arizona State University（亚利桑那州立大学物质、运输与能源工程学院）
In the trial-and-error mechanism of reinforcement learning (RL), a notorious contradiction arises when we expect to learn a safe policy: how to learn a safe policy without enough data and prior model about the dangerous region? Existing methods mostly use the posterior penalty for dangerous actions, which means that the agent is not penalized until experiencing danger. This fact causes that the agent cannot learn a zero-violation policy even after convergence. Otherwise, it would not receive any penalty and lose the knowledge about danger. In this paper, we propose the safe set actor-critic (S…
Reinforcement Learning for Exponential Utility: Algorithms and Convergence in Discounted MDPs
Computer Science and Automation, Indian Institute of Science, Bengaluru（印度科学研究院计算机科学与自动化部门，班加罗尔）；Computer Science and Engineering, Indian Institute of Technology Madras, Chennai（印度理工学院马德拉斯分校计算机科学与工程系，钦奈）；Tata Consultancy Services Limited, Hyderabad（海得拉巴市塔塔咨询公司）
Convergence of Two-Timescale Markovian Stochastic Approximations with Applications in Reinforcement Learning
Department of Computer Science, University of Virginia, Charlottesville, VA, USA（弗吉尼亚大学计算机科学系）；Data Science Lab, MIT, Cambridge, MA, USA（麻省理工学院数据科学实验室）；Mitch Daniels School of Business, Purdue University, West Lafayette, IN, USA（普渡大学米切尔丹尼尔斯商学院）；Division Office Physics, Math and Astronomy, California Institute of Technology, Pasadena, CA, USA（加州理工学院物理、数学和天文学分校）
Reinforcement learning with non-ergodic reward increments: robustness via ergodicity transformations
Aalto University（阿尔托大学）；University of Maryland（马里兰大学）；University of Warwick（华威大学）；London Mathematical Laboratory（伦敦数学实验室）；Uppsala University（乌普萨拉大学）；Santa Fe Institute（圣达菲研究所）
Arizona State University（亚利桑那州立大学）；University of Texas at Austin（德克萨斯大学奥斯汀分校）；Max Planck Institute for Software Systems（马克斯·普朗克软件系统研究所）
Despite the fact that deep reinforcement learning (RL) has surpassed human-level performances in various tasks, it still has several fundamental challenges. First, most RL methods require intensive data from the exploration of the environment to achieve satisfactory performance. Second, the use of neural networks in RL renders it hard to interpret the internals of the system in a way that humans can understand. To address these two challenges, we propose a framework that enables an RL agent to reason over its exploration process and distill high-level knowledge for effectively guiding its futu…
Policy Gradient Methods for Non-Markovian Reinforcement Learning
Department of Computer Science and Automation, Indian Institute of Science（印度科学研究院计算机科学与自动化系）；Department of Electrical Engineering, Stanford University（斯坦福大学电气工程系）；Department of Electrical and Electronics Engineering, Nanyang Technological University（南洋理工大学电气与电子工程系）；CMAP, CNRS, École polytechnique, Institut Polytechnique d́e Paris（巴黎理工学院先进材料与工艺中心、国家科学研究中心、巴黎理工学院）
Aalto University（阿尔托大学）；Delft University of Technology（代尔夫特理工大学）；University of Manchester（曼彻斯特大学）
Centaurs are half-human, half-AI decision-makers where the AI's goal is to complement the human. To do so, the AI must be able to recognize the goals and constraints of the human and have the means to help them. We present a novel formulation of the interaction between the human and the AI as a sequential game where the agents are modelled using Bayesian best-response models. We show that in this case the AI's problem of helping bounded-rational humans make better decisions reduces to a Bayes-adaptive POMDP. In our simulated experiments, we consider an instantiation of our framework for humans…
New York University Shanghai（上海纽约大学）；New York University（纽约大学）
A simple and natural algorithm for reinforcement learning (RL) is Monte Carlo Exploring Starts (MCES), where the Q-function is estimated by averaging the Monte Carlo returns, and the policy is improved by choosing actions that maximize the current estimate of the Q-function. Exploration is performed by "exploring starts", that is, each episode begins with a randomly chosen state and action, and then follows the current policy to the terminal state. In the classic book on RL by Sutton & Barto (2018), it is stated that establishing convergence for the MCES algorithm is one of the most important …
