Learning with Forgetting: an Approach to Achieve Adaptive Neural Networks

Much work on modelling pattern recognition by AI systems has focused on the stability and plasticity of a system’s ongoing response to novel inputs. This paper discusses a general learning mechanism for ART2 neural networks, which incorporates forgetting by long-term memory trace decay. Such approach enables the system to reuse memory resources and adapt to a huge in diversity or continually changing environment. Arguably the proposed learning mechanism can be considered as biologically plausible. It demonstrates features that seem to be similar to some basic characteristics of the phenomenon forgetting in the biological neural systems. Discussed experiments and simulations demonstrate some features of the learning mechanism.