Applying Policy Iteration for Training Recurrent Neural Networks

Simple item page
dc.creatorSzita, I.
dc.creatorLorincz, A.
dc.date2004-10-02
dc.date.accessioned2026-07-07T03:21:49Z
dc.date.available2026-07-07T03:21:49Z
dc.descriptionRecurrent 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.
dc.descriptionSupplementary material. 17 papes, 1 figure
dc.identifierhttps://arxiv.org/abs/cs/0410004
dc.identifierhttp://arxiv.org/abs/cs/0410004
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/32354
dc.subjectArtificial Intelligence
dc.subjectMachine Learning
dc.subjectNeural and Evolutionary Computing
dc.titleApplying Policy Iteration for Training Recurrent Neural Networks
dc.typetext

Files

Collections

arXiv