Results 1  10
of
95
Between MDPs and SemiMDPs: A Framework for Temporal Abstraction in Reinforcement Learning
 Artificial Intelligence
, 1999
"... 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 ..."
Abstract

Cited by 426 (29 self)
 Add to MetaCart
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 optionsclosedloop 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 Qlearning.
A Stochastic Model of HumanMachine Interaction for learning dialog Strategies
 IEEE Transactions on Speech and Audio Processing
, 2000
"... 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 ..."
Abstract

Cited by 167 (4 self)
 Add to MetaCart
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
Recent advances in hierarchical reinforcement learning
, 2003
"... A preliminary unedited version of this paper was incorrectly published as part of Volume ..."
Abstract

Cited by 161 (23 self)
 Add to MetaCart
A preliminary unedited version of this paper was incorrectly published as part of Volume
Infinitehorizon policygradient estimation
 Journal of Artificial Intelligence Research
, 2001
"... Gradientbased approaches to direct policy search in reinforcement learning have received much recent attention as a means to solve problems of partial observability and to avoid some of the problems associated with policy degradation in valuefunction methods. In this paper we introduce � � , a si ..."
Abstract

Cited by 153 (5 self)
 Add to MetaCart
Gradientbased approaches to direct policy search in reinforcement learning have received much recent attention as a means to solve problems of partial observability and to avoid some of the problems associated with policy degradation in valuefunction methods. In this paper we introduce � � , a simulationbased algorithm for generating a biased estimate of the gradient of the average reward in Partially Observable Markov Decision Processes ( � s) controlled by parameterized stochastic policies. A similar algorithm was proposed by Kimura, Yamamura, and Kobayashi (1995). The algorithm’s chief advantages are that it requires storage of only twice the number of policy parameters, uses one free parameter � � (which has a natural interpretation in terms of biasvariance tradeoff), and requires no knowledge of the underlying state. We prove convergence of � � , and show how the correct choice of the parameter is related to the mixing time of the controlled �. We briefly describe extensions of � � to controlled Markov chains, continuous state, observation and control spaces, multipleagents, higherorder derivatives, and a version for training stochastic policies with internal states. In a companion paper (Baxter, Bartlett, & Weaver, 2001) we show how the gradient estimates generated by � � can be used in both a traditional stochastic gradient algorithm and a conjugategradient procedure to find local optima of the average reward. 1.
MachineLearning Research  Four Current Directions
"... 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 super ..."
Abstract

Cited by 114 (1 self)
 Add to MetaCart
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.
Reinforcement Learning In Continuous Time and Space
 Neural Computation
, 2000
"... This paper presents a reinforcement learning framework for continuoustime dynamical systems without a priori discretization of time, state, and action. Based on the HamiltonJacobiBellman (HJB) equation for infinitehorizon, discounted reward problems, we derive algorithms for estimating value f ..."
Abstract

Cited by 112 (5 self)
 Add to MetaCart
This paper presents a reinforcement learning framework for continuoustime dynamical systems without a priori discretization of time, state, and action. Based on the HamiltonJacobiBellman (HJB) equation for infinitehorizon, discounted reward problems, we derive algorithms for estimating value functions and for improving policies with the use of function approximators. The process of value function estimation is formulated as the minimization of a continuoustime form of the temporal difference (TD) error. Update methods based on backward Euler approximation and exponential eligibility traces are derived and their correspondences with the conventional residual gradient, TD(0), and TD() algorithms are shown. For policy improvement, two methods, namely, a continuous actorcritic method and a valuegradient based greedy policy, are formulated. As a special case of the latter, a nonlinear feedback control law using the value gradient and the model of the input gain is derived....
Nash QLearning for GeneralSum Stochastic Games
 JOURNAL OF MACHINE LEARNING RESEARCH
, 2003
"... We extend Qlearning to a noncooperative multiagent context, using the framework of generalsum stochastic games. A learning agent maintains Qfunctions over joint actions, and performs updates based on assuming Nash equilibrium behavior over the current Qvalues. This learning protocol provably conv ..."
Abstract

Cited by 108 (0 self)
 Add to MetaCart
We extend Qlearning to a noncooperative multiagent context, using the framework of generalsum stochastic games. A learning agent maintains Qfunctions over joint actions, and performs updates based on assuming Nash equilibrium behavior over the current Qvalues. This learning protocol provably converges given certain restrictions on the stage games (defined by Qvalues) that arise during learning. Experiments with a pair of twoplayer grid games suggest that such restrictions on the game structure are not necessarily required. Stage games encountered during learning in both grid environments violate the conditions. However, learning consistently converges in the first grid game, which has a unique equilibrium Qfunction, but sometimes fails to converge in the second, which has three different equilibrium Qfunctions. In a comparison of offline learning performance in both games, we find agents are more likely to reach a joint optimal path with Nash Qlearning than with a singleagent Qlearning method. When at least one agent adopts Nash Qlearning, the performance of both agents is better than using singleagent Qlearning. We have also implemented an online version of Nash Qlearning that balances exploration with exploitation, yielding improved performance.
KernelBased Reinforcement Learning
 Machine Learning
, 1999
"... We present a kernelbased approach to reinforcement learning that overcomes the stability problems of temporaldifference learning in continuous statespaces. First, our algorithm converges to a unique solution of an approximate Bellman's equation regardless of its initialization values. Second, the ..."
Abstract

Cited by 102 (1 self)
 Add to MetaCart
We present a kernelbased approach to reinforcement learning that overcomes the stability problems of temporaldifference learning in continuous statespaces. First, our algorithm converges to a unique solution of an approximate Bellman's equation regardless of its initialization values. Second, the method is consistent in the sense that the resulting policy converges asymptotically to the optimal policy. Parametric value function estimates such as neural networks do not possess this property. Our kernelbased approach also allows us to show that the limiting distribution of the value function estimate is a Gaussian process. This information is useful in studying the biasvariance tradeo in reinforcement learning. We find that all reinforcement learning approaches to estimating the value function, parametric or nonparametric, are subject to a bias. This bias is typically larger in reinforcement learning than in a comparable regression problem.
LeastSquares Temporal Difference Learning
 In Proceedings of the Sixteenth International Conference on Machine Learning
, 1999
"... TD() is a popular family of algorithms for approximate policy evaluation in large MDPs. TD() works by incrementally updating the value function after each observed transition. It has two major drawbacks: it makes inefficient use of data, and it requires the user to manually tune a stepsize schedule ..."
Abstract

Cited by 95 (0 self)
 Add to MetaCart
TD() is a popular family of algorithms for approximate policy evaluation in large MDPs. TD() works by incrementally updating the value function after each observed transition. It has two major drawbacks: it makes inefficient use of data, and it requires the user to manually tune a stepsize schedule for good performance. For the case of linear value function approximations and = 0, the LeastSquares TD (LSTD) algorithm of Bradtke and Barto (Bradtke and Barto, 1996) eliminates all stepsize parameters and improves data efficiency. This paper extends Bradtke and Barto's work in three significant ways. First, it presents a simpler derivation of the LSTD algorithm. Second, it generalizes from = 0 to arbitrary values of ; at the extreme of = 1, the resulting algorithm is shown to be a practical formulation of supervised linear regression. Third, it presents a novel, intuitive interpretation of LSTD as a modelbased reinforcement learning technique.
Technical update: Leastsquares temporal difference learning
 Machine Learning
, 2002
"... Abstract. TD(λ) is a popular family of algorithms for approximate policy evaluation in large MDPs. TD(λ) works by incrementally updating the value function after each observed transition. It has two major drawbacks: it may make inefficient use of data, and it requires the user to manually tune a ste ..."
Abstract

Cited by 89 (2 self)
 Add to MetaCart
Abstract. TD(λ) is a popular family of algorithms for approximate policy evaluation in large MDPs. TD(λ) works by incrementally updating the value function after each observed transition. It has two major drawbacks: it may make inefficient use of data, and it requires the user to manually tune a stepsize schedule for good performance. For the case of linear value function approximations and λ = 0, the LeastSquares TD (LSTD) algorithm of Bradtke and Barto (1996, Machine learning, 22:1–3, 33–57) eliminates all stepsize parameters and improves data efficiency. This paper updates Bradtke and Barto’s work in three significant ways. First, it presents a simpler derivation of the LSTD algorithm. Second, it generalizes from λ = 0 to arbitrary values of λ; at the extreme of λ = 1, the resulting new algorithm is shown to be a practical, incremental formulation of supervised linear regression. Third, it presents a novel and intuitive interpretation of LSTD as a modelbased reinforcement learning technique.