This paper surveys the field of reinforcement learning from a computer-science perspective. It is written to be accessible to researchers familiar with machine learning. Both the historical basis of the field and a broad selection of current work are summarized. Reinforcement learning is the problem faced by an agent that learns behavior through trial-and-error interactions with a dynamic environment. The work described here has a resemblance to work in psychology, but differs considerably in the details and in the use of the word "reinforcement." The paper discusses central issues of reinforcement learning, including trading off exploration and exploitation, establishing the foundations of the field via Markov decision theory, learning from delayed reinforcement, constructing empirical models to accelerate learning, making use of generalization and hierarchy, and coping with hidden state. It concludes with a survey of some implemented systems and an assessment of the practical utility of current methods for reinforcement learning.

Learning, planning, and representing knowledge at multiple levels of temporal abstraction are key, longstanding challenges for AI. In this paper we consider how these challenges can be addressed within the mathematical framework of reinforcement learning and Markov decision processes (MDPs). We extend the usual notion of action in this framework to include options---closed-loop policies for taking action over a period of time. Examples of options include picking up an object, going to lunch, and traveling to a distant city, as well as primitive actions such as muscle twitches and joint torques. Overall, we show that options enable temporally abstract knowledge and action to be included in the reinforcement learning framework in a natural and general way. In particular, we show that options may be used interchangeably with primitive actions in planning methods such as dynamic programming and in learning methods such as Q-learning.

On large problems, reinforcement learning systems must use parameterized function approximators such as neural networks in order to generalize between similar situations and actions. In these cases there are no strong theoretical results on the accuracy of convergence, and computational results have been mixed. In particular, Boyan and Moore reported at last year's meeting a series of negative results in attempting to apply dynamic programming together with function approximation to simple control problems with continuous state spaces. In this paper, we present positive results for all the control tasks they attempted, and for one that is significantly larger. The most important differences are that we used sparse-coarse-coded function approximators (CMACs) whereas they used mostly global function approximators, and that we learned online whereas they learned offline. Boyan and Moore and others have suggested that the problems they encountered could be solved by using actual outcomes (...

Abstract—In this paper, we propose a quantitative model for dialog systems that can be used for learning the dialog strategy. We claim that the problem of dialog design can be formalized as an optimization problem with an objective function reflecting different dialog dimensions relevant for a given application. We also show that any dialog system can be formally described as a sequential decision process in terms of its state space, action set, and strategy. With additional assumptions about the state transition probabilities and cost assignment, a dialog system can be mapped to a stochastic model known as Markov decision process (MDP). A variety of data driven algorithms for finding the optimal strategy (i.e., the one that optimizes the criterion) is available within the MDP framework, based on reinforcement learning. For an effective use of the available training data we propose a combination of supervised and reinforcement learning: the supervised learning is used to estimate a model of the user, i.e., the MDP parameters that quantify the user’s behavior. Then a reinforcement learning algorithm is used to estimate the optimal strategy while the system interacts with the simulated user. This approach is tested for learning the strategy in an air travel information system (ATIS) task. The experimental results we present in this paper show that it is indeed possible to find a simple criterion, a state space representation, and a simulated user parameterization in order to automatically learn a relatively complex dialog behavior, similar to one that was heuristically designed by several research groups. Index Terms—Dialog systems, Markov decision process, reinforcement learning, sequential decision process, speech, spoken

In cellular telephone systems, an important problem is to dynamically allocate the communication resource (channels) so as to maximize service in a stochastic caller environment. This problem is naturally formulated as a dynamic programming problem and we use a reinforcement learning (RL) method to find dynamic channel allocation policies that are better than previous heuristic solutions. The policies obtained perform well for a broad variety of call traffic patterns. We present results on a large cellular system with approximately 49^49 states.

The curse of dimensionality gives rise to prohibitive computational requirements that render infeasible the exact solution of large-scale stochastic control problems. We study an efficient method based on linear programming for approximating solutions to such problems. The approach “fits ” a linear combination of pre-selected basis functions to the dynamic programming cost-to-go function. We develop error bounds that offer performance guarantees and also guide the selection of both basis functions and “state-relevance weights ” that influence quality of the approximation. Experimental results in the domain of queueing network control provide empirical support for the methodology. (Dynamic programming/optimal control: approximations/large-scale problems. Queues, algorithms: control of queueing networks.)

Machine Learning research has been making great progress in many directions. This article summarizes four of these directions and discusses some current open problems. The four directions are (a) improving classification accuracy by learning ensembles of classifiers, (b) methods for scaling up supervised learning algorithms, (c) reinforcement learning, and (d) learning complex stochastic models.

Following Tesauro's work on TD-Gammon, we used a 4000 parameter feed-forward neural network to develop a competitive backgammon evaluation function. Play proceeds by a roll of the dice, application of the network to all legal moves, and choosing the move with the highest evaluation. However, no back-propagation, reinforcement or temporal difference learning methods were employed. Instead we apply simple hill-climbing in a relative fitness environment. We start with an initial champion of all zero weights and proceed simply by playing the current champion network against a slightly mutated challenger and changing weights if the challenger wins. Surprisingly, this worked rather well. We investigate how the peculiar dynamics of this domain enabled a previously discarded weak method to succeed, by preventing suboptimal equilibria in a "meta-game" of self-learning. Keywords: coevolution, backgammon, reinforcement, temporal difference learning, self-learning Running Head: CO-EVOLUTIONARY LEA...

This dissertation investigates the use of hierarchy and problem decomposition as a means of solving large, stochastic, sequential decision problems. These problems are framed as Markov decision problems (MDPs). The new technical content of this dissertation begins with a discussion of the concept of temporal abstraction. Temporal abstraction is shown to be equivalent to the transformation of a policy defined over a region of an MDP to an action in a semi-Markov decision problem (SMDP). Several algorithms are presented for performing this transformation efficiently. This dissertation introduces the HAM method for generating hierarchical, temporally abstract actions. This method permits the partial specification of abstract actions in a way that corresponds to an abstract plan or strategy. Abstr...

Designing the dialogue policy of a spoken dialogue system involves many nontrivial choices. This paper presents a reinforcement learning approach for automatically optimizing a dialogue policy, which addresses the technical challenges in applying reinforcement learning to a working dialogue system with human users. We report on the design, construction and empirical evaluation of NJFun, an experimental spoken dialogue system that provides users with access to information about fun things to do in New Jersey. Our results show that by optimizing its performance via reinforcement learning, NJFun measurably improves system performance.