AUTHOR=Rahman Mustafizur , Bose Subhankar , Chakrabartty Shantanu 

TITLE=On-device synaptic memory consolidation using Fowler-Nordheim quantum-tunneling

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 16 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.1050585

DOI=10.3389/fnins.2022.1050585

ISSN=1662-453X

ABSTRACT=For artificial synapses whose strengths are assumed to be bounded and can only be updated with finite precision, achieving an optimal memory consolidation characteristic using classical physics leads to synaptic models that are too complex to be scaled in-silico. Here we report that a relatively simple differential device that operates using the physics of Fowler-Nordheim (FN) quantum-mechanical tunneling can achieve tunable memory consolidation characteristics. We show that compared to other physical implementations of complex synapses, the operation of the FN-synapse is near-optimal in terms of the synaptic lifetime and the consolidation properties can be easily tuned to achieve different plasticity-stability trade-offs. With an energy footprint of femtojoules per synaptic update, we believe that the proposed FN-synapse provides an ultra-energy-efficient approach for implementing both synaptic memory consolidation and continual learning on a physical device.