Results 11  20
of
313
Experiments with Reinforcement Learning in Problems with Continuous State and Action Spaces
, 1996
"... A key element in the solution of reinforcement learning problems is the value function. The purpose of this function is to measure the longterm utility or value of any given state and it is important because an agent can use it to decide what to do next. A common problem in reinforcement learning w ..."
Abstract

Cited by 115 (6 self)
 Add to MetaCart
A key element in the solution of reinforcement learning problems is the value function. The purpose of this function is to measure the longterm utility or value of any given state and it is important because an agent can use it to decide what to do next. A common problem in reinforcement learning when applied to systems having continuous states and action spaces is that the value function must operate with a domain consisting of realvalued variables, which means that it should be able to represent the value of infinitely many state and action pairs. For this reason, function approximators are used to represent the value function when a closeform solution of the optimal policy is not available. In this paper, we extend a previously proposed reinforcement learning algorithm so that it can be used with function approximators that generalize the value of individual experiences across both, state and action spaces. In particular, we discuss the benefits of using sparse coarsecoded funct...
Optimal Stopping of Markov Processes: Hilbert Space Theory, Approximation Algorithms, and an Application to Pricing HighDimensional Financial Derivatives
 IEEE Transactions on Automatic Control
, 1997
"... We develop a theory characterizing optimal stopping times for discretetime ergodic Markov processes with discounted rewards. The theory differs from prior work by its view of perstage and terminal reward functions as elements of a certain Hilbert space. In addition to a streamlined analysis establ ..."
Abstract

Cited by 114 (6 self)
 Add to MetaCart
We develop a theory characterizing optimal stopping times for discretetime ergodic Markov processes with discounted rewards. The theory differs from prior work by its view of perstage and terminal reward functions as elements of a certain Hilbert space. In addition to a streamlined analysis establishing existence and uniqueness of a solution to Bellman's equation, this approach provides an elegant framework for the study of approximate solutions. In particular, we propose a stochastic approximation algorithm that tunes weights of a linear combination of basis functions in order to approximate a value function. We prove that this algorithm converges (almost surely) and that the limit of convergence has some desirable properties. We discuss how variations on this line of analysis can be used to develop similar results for other classes of optimal stopping problems, including those involving independent increment processes, finite horizons, and twoplayer zerosum games. We illustrate...
Rationality and intelligence
 Artificial Intelligence
, 1997
"... The longterm goal of our field is the creation and understanding of intelligence. Productive research in AI, both practical and theoretical, benefits from a notion of intelligence that is precise enough to allow the cumulative development of robust systems and general results. This paper outlines a ..."
Abstract

Cited by 105 (1 self)
 Add to MetaCart
The longterm goal of our field is the creation and understanding of intelligence. Productive research in AI, both practical and theoretical, benefits from a notion of intelligence that is precise enough to allow the cumulative development of robust systems and general results. This paper outlines a gradual evolution in our formal conception of intelligence that brings it closer to our informal conception and simultaneously reduces the gap between theory and practice. 1 Artificial Intelligence AI is a field in which the ultimate goal has often been somewhat illdefined and subject to dispute. Some researchers aim to emulate human cognition, others aim at the creation of
Protovalue functions: A laplacian framework for learning representation and control in markov decision processes
 Journal of Machine Learning Research
, 2006
"... This paper introduces a novel spectral framework for solving Markov decision processes (MDPs) by jointly learning representations and optimal policies. The major components of the framework described in this paper include: (i) A general scheme for constructing representations or basis functions by d ..."
Abstract

Cited by 92 (10 self)
 Add to MetaCart
This paper introduces a novel spectral framework for solving Markov decision processes (MDPs) by jointly learning representations and optimal policies. The major components of the framework described in this paper include: (i) A general scheme for constructing representations or basis functions by diagonalizing symmetric diffusion operators (ii) A specific instantiation of this approach where global basis functions called protovalue functions (PVFs) are formed using the eigenvectors of the graph Laplacian on an undirected graph formed from state transitions induced by the MDP (iii) A threephased procedure called representation policy iteration comprising of a sample collection phase, a representation learning phase that constructs basis functions from samples, and a final parameter estimation phase that determines an (approximately) optimal policy within the (linear) subspace spanned by the (current) basis functions. (iv) A specific instantiation of the RPI framework using leastsquares policy iteration (LSPI) as the parameter estimation method (v) Several strategies for scaling the proposed approach to large discrete and continuous state spaces, including the Nyström extension for outofsample interpolation of eigenfunctions, and the use of Kronecker sum factorization to construct compact eigenfunctions in product spaces such as factored MDPs (vi) Finally, a series of illustrative discrete and continuous control tasks, which both illustrate the concepts and provide a benchmark for evaluating the proposed approach. Many challenges remain to be addressed in scaling the proposed framework to large MDPs, and several elaboration of the proposed framework are briefly summarized at the end.
Least Squares Policy Evaluation Algorithms With Linear Function Approximation
 Theory and Applications
, 2002
"... We consider policy evaluation algorithms within the context of infinitehorizon dynamic programming problems with discounted cost. We focus on discretetime dynamic systems with a large number of states, and we discuss two methods, which use simulation, temporal differences, and linear cost function ..."
Abstract

Cited by 92 (12 self)
 Add to MetaCart
We consider policy evaluation algorithms within the context of infinitehorizon dynamic programming problems with discounted cost. We focus on discretetime dynamic systems with a large number of states, and we discuss two methods, which use simulation, temporal differences, and linear cost function approximation. The first method is a new gradientlike algorithm involving leastsquares subproblems and a diminishing stepsize, which is based on the #policy iteration method of Bertsekas and Ioffe. The second method is the LSTD(#) algorithm recently proposed by Boyan, which for # =0coincides with the linear leastsquares temporaldifference algorithm of Bradtke and Barto. At present, there is only a convergence result by Bradtke and Barto for the LSTD(0) algorithm. Here, we strengthen this result by showing the convergence of LSTD(#), with probability 1, for every # [0, 1].
Regularization and feature selection in leastsquares temporal difference learning
, 2009
"... We consider the task of reinforcement learning with linear value function approximation. Temporal difference algorithms, and in particular the LeastSquares Temporal Difference (LSTD) algorithm, provide a method for learning the parameters of the value function, but when the number of features is la ..."
Abstract

Cited by 80 (1 self)
 Add to MetaCart
(Show Context)
We consider the task of reinforcement learning with linear value function approximation. Temporal difference algorithms, and in particular the LeastSquares Temporal Difference (LSTD) algorithm, provide a method for learning the parameters of the value function, but when the number of features is large this algorithm can overfit to the data and is computationally expensive. In this paper, we propose a regularization framework for the LSTD algorithm that overcomes these difficulties. In particular, we focus on the case of l1 regularization, which is robust to irrelevant features and also serves as a method for feature selection. Although the l1 regularized LSTD solution cannot be expressed as a convex optimization problem, we present an algorithm similar to the Least Angle Regression (LARS) algorithm that can efficiently compute the optimal solution. Finally, we demonstrate the performance of the algorithm experimentally.
Automatic basis function construction for approximate dynamic programming and reinforcement learning
 In Cohen and Moore (2006
, 2006
"... We address the problem of automatically constructing basis functions for linear approximation of the value function of a Markov Decision Process (MDP). Our work builds on results by Bertsekas and Castañon (1989) who proposed a method for automatically aggregating states to speed up value iteration. ..."
Abstract

Cited by 77 (4 self)
 Add to MetaCart
We address the problem of automatically constructing basis functions for linear approximation of the value function of a Markov Decision Process (MDP). Our work builds on results by Bertsekas and Castañon (1989) who proposed a method for automatically aggregating states to speed up value iteration. We propose to use neighborhood component analysis (Goldberger et al., 2005), a dimensionality reduction technique created for supervised learning, in order to map a highdimensional state space to a lowdimensional space, based on the Bellman error, or on the temporal difference (TD) error. We then place basis function in the lowerdimensional space. These are added as new features for the linear function approximator. This approach is applied to a highdimensional inventory control problem. 1.
Bayes Meets Bellman: The Gaussian Process Approach to Temporal Difference Learning
 Proc. of the 20th International Conference on Machine Learning
, 2003
"... We present a novel Bayesian approach to the problem of value function estimation in continuous state spaces. We de ne a probabilistic generative model for the value function by imposing a Gaussian prior over value functions and assuming a Gaussian noise model. ..."
Abstract

Cited by 76 (8 self)
 Add to MetaCart
We present a novel Bayesian approach to the problem of value function estimation in continuous state spaces. We de ne a probabilistic generative model for the value function by imposing a Gaussian prior over value functions and assuming a Gaussian noise model.
Incremental Natural ActorCritic Algorithms
"... We present four new reinforcement learning algorithms based on actorcritic and naturalgradient ideas, and provide their convergence proofs. Actorcritic reinforcement learning methods are online approximations to policy iteration in which the valuefunction parameters are estimated using temporal ..."
Abstract

Cited by 75 (8 self)
 Add to MetaCart
We present four new reinforcement learning algorithms based on actorcritic and naturalgradient ideas, and provide their convergence proofs. Actorcritic reinforcement learning methods are online approximations to policy iteration in which the valuefunction parameters are estimated using temporal difference learning and the policy parameters are updated by stochastic gradient descent. Methods based on policy gradients in this way are of special interest because of their compatibility with function approximation methods, which are needed to handle large or infinite state spaces. The use of temporal difference learning in this way is of interest because in many applications it dramatically reduces the variance of the gradient estimates. The use of the natural gradient is of interest because it can produce better conditioned parameterizations and has been shown to further reduce variance in some cases. Our results extend prior twotimescale convergence results for actorcritic methods by Konda and Tsitsiklis by using temporal difference learning in the actor and by incorporating natural gradients, and they extend prior empirical studies of natural actorcritic methods by Peters, Vijayakumar and Schaal by providing the first convergence proofs and the first fully incremental algorithms. 1
Basis function adaptation in temporal difference reinforcement learning
 Annals of Operations Research
, 2005
"... Reinforcement Learning (RL) is an approach for solving complex multistage decision problems that fall under the general framework of Markov Decision Problems (MDPs), with possibly unknown parameters. Function approximation is essential for problems with a large state space, as it facilitates compact ..."
Abstract

Cited by 73 (4 self)
 Add to MetaCart
Reinforcement Learning (RL) is an approach for solving complex multistage decision problems that fall under the general framework of Markov Decision Problems (MDPs), with possibly unknown parameters. Function approximation is essential for problems with a large state space, as it facilitates compact representation and enables generalization. Linear approximation architectures (where the adjustable parameters are the weights of prefixed basis functions) have recently gained prominence due to efficient algorithms and convergence guarantees. Nonetheless, an appropriate choice of basis function is important for the success of the algorithm. In the present paper we examine methods for adapting the basis function during the learning process in the context of evaluating the value function under a fixed control policy. Using the Bellman approximation error as an optimization criterion, we optimize the weights of the basis function while simultaneously adapting the (nonlinear) basis function parameters. We present two algorithms for this problem. The first uses a gradientbased approach and the second applies the Cross Entropy method. The performance of the proposed algorithms is evaluated and compared in simulations.