Technical Note Q-Learning
01 May 1992-Machine Learning (Kluwer Academic Publishers-Plenum Publishers)-Vol. 8, Iss: 3, pp 279-292
TL;DR: In this article, it is shown that Q-learning converges to the optimum action-values with probability 1 so long as all actions are repeatedly sampled in all states and the action values are represented discretely.
Abstract: Q-learning (Watkins, 1989) is a simple way for agents to learn how to act optimally in controlled Markovian domains. It amounts to an incremental method for dynamic programming which imposes limited computational demands. It works by successively improving its evaluations of the quality of particular actions at particular states. This paper presents and proves in detail a convergence theorem for Q,-learning based on that outlined in Watkins (1989). We show that Q-learning converges to the optimum action-values with probability 1 so long as all actions are repeatedly sampled in all states and the action-values are represented discretely. We also sketch extensions to the cases of non-discounted, but absorbing, Markov environments, and where many Q values can be changed each iteration, rather than just one.
Citations
More filters
TL;DR: This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.
Abstract: The theory of reinforcement learning provides a normative account, deeply rooted in psychological and neuroscientific perspectives on animal behaviour, of how agents may optimize their control of an environment. To use reinforcement learning successfully in situations approaching real-world complexity, however, agents are confronted with a difficult task: they must derive efficient representations of the environment from high-dimensional sensory inputs, and use these to generalize past experience to new situations. Remarkably, humans and other animals seem to solve this problem through a harmonious combination of reinforcement learning and hierarchical sensory processing systems, the former evidenced by a wealth of neural data revealing notable parallels between the phasic signals emitted by dopaminergic neurons and temporal difference reinforcement learning algorithms. While reinforcement learning agents have achieved some successes in a variety of domains, their applicability has previously been limited to domains in which useful features can be handcrafted, or to domains with fully observed, low-dimensional state spaces. Here we use recent advances in training deep neural networks to develop a novel artificial agent, termed a deep Q-network, that can learn successful policies directly from high-dimensional sensory inputs using end-to-end reinforcement learning. We tested this agent on the challenging domain of classic Atari 2600 games. We demonstrate that the deep Q-network agent, receiving only the pixels and the game score as inputs, was able to surpass the performance of all previous algorithms and achieve a level comparable to that of a professional human games tester across a set of 49 games, using the same algorithm, network architecture and hyperparameters. This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.
23,074 citations
TL;DR: This historical survey compactly summarizes relevant work, much of it from the previous millennium, review deep supervised learning, unsupervised learning, reinforcement learning & evolutionary computation, and indirect search for short programs encoding deep and large networks.
14,635 citations
TL;DR: This Review looks at some key brain theories in the biological and physical sciences from the free-energy perspective, suggesting that several global brain theories might be unified within a free- energy framework.
Abstract: A free-energy principle has been proposed recently that accounts for action, perception and learning. This Review looks at some key brain theories in the biological (for example, neural Darwinism) and physical (for example, information theory and optimal control theory) sciences from the free-energy perspective. Crucially, one key theme runs through each of these theories — optimization. Furthermore, if we look closely at what is optimized, the same quantity keeps emerging, namely value (expected reward, expected utility) or its complement, surprise (prediction error, expected cost). This is the quantity that is optimized under the free-energy principle, which suggests that several global brain theories might be unified within a free-energy framework.
4,866 citations
10 Jul 1994
TL;DR: A Q-learning-like algorithm for finding optimal policies and its application to a simple two-player game in which the optimal policy is probabilistic is demonstrated.
Abstract: In the Markov decision process (MDP) formalization of reinforcement learning, a single adaptive agent interacts with an environment defined by a probabilistic transition function. In this solipsis-tic view, secondary agents can only be part of the environment and are therefore fixed in their behavior. The framework of Markov games allows us to widen this view to include multiple adaptive agents with interacting or competing goals. This paper considers a step in this direction in which exactly two agents with diametrically opposed goals share an environment. It describes a Q-learning-like algorithm for finding optimal policies and demonstrates its application to a simple two-player game in which the optimal policy is probabilistic.
2,643 citations
TL;DR: This tutorial presents the CE methodology, the basic algorithm and its modifications, and discusses applications in combinatorial optimization and machine learning.
Abstract: The cross-entropy (CE) method is a new generic approach to combinatorial and multi-extremal optimization and rare event simulation. The purpose of this tutorial is to give a gentle introduction to the CE method. We present the CE methodology, the basic algorithm and its modifications, and discuss applications in combinatorial optimization and machine learning.
2,367 citations
References
More filters
TL;DR: This article introduces a class of incremental learning procedures specialized for prediction – that is, for using past experience with an incompletely known system to predict its future behavior – and proves their convergence and optimality for special cases and relation to supervised-learning methods.
Abstract: This article introduces a class of incremental learning procedures specialized for prediction – that is, for using past experience with an incompletely known system to predict its future behavior. Whereas conventional prediction-learning methods assign credit by means of the difference between predicted and actual outcomes, the new methods assign credit by means of the difference between temporally successive predictions. Although such temporal-difference methods have been used in Samuel's checker player, Holland's bucket brigade, and the author's Adaptive Heuristic Critic, they have remained poorly understood. Here we prove their convergence and optimality for special cases and relate them to supervised-learning methods. For most real-world prediction problems, temporal-difference methods require less memory and less peak computation than conventional methods and they produce more accurate predictions. We argue that most problems to which supervised learning is currently applied are really prediction problems of the sort to which temporal-difference methods can be applied to advantage.
4,803 citations
TL;DR: This paper compares eight reinforcement learning frameworks: Adaptive heuristic critic (AHC) learning due to Sutton, Q-learning due to Watkins, and three extensions to both basic methods for speeding up learning and two extensions are experience replay, learning action models for planning, and teaching.
Abstract: To date, reinforcement learning has mostly been studied solving simple learning tasks. Reinforcement learning methods that have been studied so far typically converge slowly. The purpose of this work is thus two-fold: 1) to investigate the utility of reinforcement learning in solving much more complicated learning tasks than previously studied, and 2) to investigate methods that will speed up reinforcement learning.
This paper compares eight reinforcement learning frameworks: adaptive heuristic critic (AHC) learning due to Sutton, Q-learning due to Watkins, and three extensions to both basic methods for speeding up learning. The three extensions are experience replay, learning action models for planning, and teaching. The frameworks were investigated using connectionism as an approach to generalization. To evaluate the performance of different frameworks, a dynamic environment was used as a testbed. The environment is moderately complex and nondeterministic. This paper describes these frameworks and algorithms in detail and presents empirical evaluation of the frameworks.
1,691 citations
01 Jun 1990
TL;DR: This paper extends previous work with Dyna, a class of architectures for intelligent systems based on approximating dynamic programming methods, and presents and shows results for two Dyna architectures, based on Watkins's Q-learning, a new kind of reinforcement learning.
Abstract: This paper extends previous work with Dyna, a class of architectures for intelligent systems based on approximating dynamic programming methods. Dyna architectures integrate trial-and-error (reinforcement) learning and execution-time planning into a single process operating alternately on the world and on a learned model of the world. In this paper, I present and show results for two Dyna architectures. The Dyna-PI architecture is based on dynamic programming's policy iteration method and can be related to existing AI ideas such as evaluation functions and universal plans (reactive systems). Using a navigation task, results are shown for a simple Dyna-PI system that simultaneously learns by trial and error, learns a world model, and plans optimal routes using the evolving world model. The Dyna-Q architecture is based on Watkins's Q-learning, a new kind of reinforcement learning. Dyna-Q uses a less familiar set of data structures than does Dyna-PI, but is arguably simpler to implement and use. We show that Dyna-Q architectures are easy to adapt for use in changing environments.
1,592 citations