Adapting to time: why nature chose to evolve a diverse set of neurons

Authors: Karim G. Habashy, Benjamin D. Evans, Dan F. M. Goodman and Jeffrey S. Bowers

Presentation type: Talk at SNUFA 2024 online workshop (5-6 Nov 2024)

Abstract

Evolution has yielded a diverse set of neurons with varying morphologies that impact their processing of temporal information. In addition, it is known empirically that spike timing is a significant factor in neural computations. However, despite these two observations, most neural network models deal with spatially structured inputs with time-locked time steps, while mainly restricting variation to parameters like weights and biases. In this study, we investigate the relevance of adapting temporal parameters, like time constants and delays, in feedforward networks that map spatio-temporal spike patterns. In this context, we show that networks with richer potential dynamics are able to more easily and robustly learn tasks with temporal structure. We suggest that it is therefore likely that the brain uses at least some of these strategies in response to a time varying and noisy environment. More precisely, from these investigations, we demonstrate the limitations of weight-only adaption, show the advantage of incorporating temporal parameters and elaborate on the interaction between the various parameters. Finally, we discuss the advantages of adapting temporal parameters when dealing with noise in inputs and weights, which might prove useful in neuromorphic hardware design.