Applying Policy Iteration for Training Recurrent Neural Networks
Artificial Intelligence
2007-05-23 v1 Machine Learning
Neural and Evolutionary Computing
Abstract
Recurrent neural networks are often used for learning time-series data. Based on a few assumptions we model this learning task as a minimization problem of a nonlinear least-squares cost function. The special structure of the cost function allows us to build a connection to reinforcement learning. We exploit this connection and derive a convergent, policy iteration-based algorithm. Furthermore, we argue that RNN training can be fit naturally into the reinforcement learning framework.
Cite
@article{arxiv.cs/0410004,
title = {Applying Policy Iteration for Training Recurrent Neural Networks},
author = {I. Szita and A. Lorincz},
journal= {arXiv preprint arXiv:cs/0410004},
year = {2007}
}
Comments
Supplementary material. 17 papes, 1 figure