Published: 
By  Virginia Nano-Computing Group (VINO) Ghosh

Golam's paper entitled "Choose your tools carefully: a comparative evaluation of deterministic vs. stochastic and binary vs. analog neuron models for implementing emerging computing paradigms (link)" published in Frontiers in Nanotechnology. In this paper, G. Morshed et al. show a comparative analysis between analog and binary neuron models (including stochasticity) for analog time-series prediction tasks. The results suggest analog stochastic neuron (ASN) model shows better performance metrics, i.e., inference errors, generalizability and robustness, practical implementability, and memory capacity, than its binary counterpart.