Robust reservoir generation by correlation-based learning

Authors:
Tadashi Yamazaki;Shigeru Tanaka
Affiliations:
Laboratory for Motor Learning Control, RIKEN Brain Science Institute, Saitama, Japan;Laboratory for Visual Neurocomputing, RIKEN Brain Science Institute, Saitama, Japan
Venue:
Advances in Artificial Neural Systems
Year:
2009

Citing 4
Cited 0

Real-time computing without stable states: a new framework for neural computation based on perturbations

Neural Computation
Neural Modeling of an Internal Clock

Neural Computation
Editorial: Special issue on echo state networks and liquid state machines

Neural Networks
2007 Special Issue: The cerebellum as a liquid state machine

Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Reservoir computing (RC) is a new framework for neural computation. A reservoir is usually a recurrent neural network with fixed random connections. In this article, we propose an RCmodel in which the connections in the reservoir are modifiable. Specifically, we consider correlation-based learning (CBL), which modifies the connection weight between a given pair of neurons according to the correlation in their activities. We demonstrate that CBL enables the reservoir to reproduce almost the same spatiotemporal activity patterns in response to an identical input stimulus in the presence of noise. This result suggests that CBL enhances the robustness in the generation of the spatiotemporal activity pattern against noise in input signals. We apply our RC model to trace eyeblink conditioning. The reservoir bridged the gap of an interstimulus interval between the conditioned and unconditioned stimuli, and a readout neuron was able to learn and express the timed conditioned response.