<title>Adapting robot behavior to a nonstationary environment: a deeper biologically inspired model of neural processing</title>

Abstract

Biological inspiration admits to degrees. This paper describes a new neural processing algorithm inspired by a deeper understanding of the workings of real biological synapses. It is shown that a multi-time domain adaptation approach to encoding causal correlation solves the destructive interference problem encountered by more commonly used learning algorithms. It is also shown how this allows an agent to adapt to nonstationary environments in which longer-term changes in the statistical properties occur and are inherently unpredictable, yet not completely lose useful prior knowledge. Finally, it is suggested that the use of causal correlation coupled with value-based learning may provide pragmatic solutions to some other classical problems in machine learning.