Applying Policy Iteration for Training Recurrent Neural Networks

Details Applying Policy Iteration for Training Recurrent Neural Networks Applying Policy Iteration for Training Recurrent Neural Networks

2004-10-02

arxiv.org/abs/cs/0410004v1

Computer Science

Artificial Intelligence

Supplementary material. 17 papes, 1 figure

Scientific paper

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.

Affiliated with

Also associated with

No associations

Rating

Applying Policy Iteration for Training Recurrent Neural Networks does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Applying Policy Iteration for Training Recurrent Neural Networks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Applying Policy Iteration for Training Recurrent Neural Networks will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-351585