Simultaneous self-organization of place and direction selectivity in a neural model of self-localization

Abstract

We present a novel approach to the problem of generating internal representations that are predictive for future stimulations of an autonomous agent. In our network laterally propagated activities provide expectancy biases for subsequent activations. Activations are driven by inputs from sensors for own actions and the environment. During learning sensory… (More)
DOI: 10.1016/S0925-2312(98)00140-4

Topics

4 Figures and Tables

Slides referencing similar topics