Experiments With Reservoir Computing On The Road Sign Problem

Title: Experiments With Reservoir Computing On The Road Sign Problem

Authors: Antonelo, Eric Aislan; Schrauwen, Benjamin; Stroobandt, Dirk

Resumo: The road sign problem is tackled in this work with Reservoir Computing (RC) networks. These networks are made of a fixed recurrent neural network where only a readout layer is trained. In the road sign problem, an agent has to decide at some point in time which action to take given relevant information gathered in the past. We show that RC can handle simple and complex T-maze tasks (which are a subdomain of the road sign problem).

Keywords: Reservoir Computing; road sign problem; T-maze; long-term memory

Pages: 6

DOI: 10.21528/CBRN2007-047

Paper as PDF: 50100047.pdf

BibTex file: 50100047.bib