Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning

This article presents a general class of associative reinforcement learning algorithms for connectionist networks containing stochastic units. These algorithms, called REINFORCE algorithms, are shown to make weight adjustments in a direction that lies along the gradient of expected reinforcement in both immediate-reinforcement tasks and certain limited forms of delayed-reinforcement tasks, and they do this without explicitly computing gradient estimates or even storing information from which such estimates could be computed. Specific examples of such algorithms are presented, some of which bear a close relationship to certain existing algorithms while others are novel but potentially interesting in their own right. Also given are results that show how such algorithms can be naturally integrated with backpropagation. We close with a brief discussion of a number of additional issues surrounding the use of such algorithms, including what is known about their limiting behaviors as well as further considerations that might be used to help develop similar but potentially more powerful reinforcement learning algorithms.

[1]  Nils J. Nilsson,et al.  Principles of Artificial Intelligence , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Richard S. Sutton,et al.  Learning and Sequential Decision Making , 1989 .

[3]  Vijaykumar Gullapalli,et al.  A stochastic reinforcement learning algorithm for learning real-valued functions , 1990, Neural Networks.

[4]  Kumpati S. Narendra,et al.  Learning automata - an introduction , 1989 .

[5]  P. Anandan,et al.  Pattern-recognizing stochastic learning automata , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[6]  Kumpati S. Narendra,et al.  An N-player sequential stochastic game with identical payoffs , 1983, IEEE Transactions on Systems, Man, and Cybernetics.

[7]  Graham C. Goodwin,et al.  Adaptive filtering prediction and control , 1984 .

[8]  Geoffrey E. Hinton,et al.  Learning and relearning in Boltzmann machines , 1986 .

[9]  P. Werbos,et al.  Beyond Regression : "New Tools for Prediction and Analysis in the Behavioral Sciences , 1974 .

[10]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[11]  K. Narendra,et al.  Decentralized learning in finite Markov chains , 1985, 1985 24th IEEE Conference on Decision and Control.

[12]  Yann LeCun,et al.  Une procedure d'apprentissage pour reseau a seuil asymmetrique (A learning scheme for asymmetric threshold networks) , 1985 .

[13]  Richard S. Sutton,et al.  Temporal credit assignment in reinforcement learning , 1984 .

[14]  M. A. L. THATHACHAR,et al.  A new approach to the design of reinforcement schemes for learning automata , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[15]  Jing Peng,et al.  Function Optimization using Connectionist Reinforcement Learning Algorithms , 1991 .

[16]  A G Barto,et al.  Learning by statistical cooperation of self-interested neuron-like computing elements. , 1985, Human neurobiology.

[17]  C. Watkins.  Learning from delayed rewards , 1989 .

[18]  Michael I. Jordan,et al.  Forward Models: Supervised Learning with a Distal Teacher , 1992, Cogn. Sci..

[19]  R. J. Williams,et al.  On the use of backpropagation in associative reinforcement learning , 1988, IEEE 1988 International Conference on Neural Networks.