Contrastive Power-Efficient Physical Learning in Resistor Networks
Abstract
The prospect of substantial reductions in the power consumption of AI is a major motivation for the development of neuromorphic hardware. Less attention has been given to the complementary research of power-efficient learning rules for such systems. Here we study self-learning physical systems trained by local learning rules based on contrastive learning. We show how the physical learning rule can be biased toward finding power-efficient solutions to learning problems, and demonstrate in simulations and laboratory experiments the emergence of a trade-off between power-efficiency and task performance.
Cite
Text
Stern et al. "Contrastive Power-Efficient Physical Learning in Resistor Networks." NeurIPS 2023 Workshops: MLNCP, 2023.Markdown
[Stern et al. "Contrastive Power-Efficient Physical Learning in Resistor Networks." NeurIPS 2023 Workshops: MLNCP, 2023.](https://mlanthology.org/neuripsw/2023/stern2023neuripsw-contrastive/)BibTeX
@inproceedings{stern2023neuripsw-contrastive,
title = {{Contrastive Power-Efficient Physical Learning in Resistor Networks}},
author = {Stern, Menachem and Dillavou, Sam and Jayaraman, Dinesh and Durian, Douglas and Liu, Andrea},
booktitle = {NeurIPS 2023 Workshops: MLNCP},
year = {2023},
url = {https://mlanthology.org/neuripsw/2023/stern2023neuripsw-contrastive/}
}