Remembrance of Tasks Past in Tunable Physical Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chatterjee, Purba, Guzman, Marcelo, Liu, Andrea J.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912746419781632
author Chatterjee, Purba
Guzman, Marcelo
Liu, Andrea J.
author_facet Chatterjee, Purba
Guzman, Marcelo
Liu, Andrea J.
contents Sequential learning in physical networks is hindered by catastrophic forgetting, where training a new task erases solutions to earlier ones. We show that we can significantly enhance memory of previous tasks by introducing a hard threshold in the learning rule, allowing only edges with sufficiently large training signals to be altered. Thresholding confines tuning to the spatial vicinity of inputs and outputs for each task, effectively partitioning the network into weakly overlapping functional regions. Using simulations of tunable resistor networks, we demonstrate that this strategy enables robust memory of multiple sequential tasks while reducing the number of edges and the overall tuning cost. Our results hint at constrained training as a simple, local, and scalable mechanism to overcome catastrophic forgetting in tunable matter.
format Preprint
id arxiv_https___arxiv_org_abs_2512_03799
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Remembrance of Tasks Past in Tunable Physical Networks
Chatterjee, Purba
Guzman, Marcelo
Liu, Andrea J.
Disordered Systems and Neural Networks
Soft Condensed Matter
Statistical Mechanics
Sequential learning in physical networks is hindered by catastrophic forgetting, where training a new task erases solutions to earlier ones. We show that we can significantly enhance memory of previous tasks by introducing a hard threshold in the learning rule, allowing only edges with sufficiently large training signals to be altered. Thresholding confines tuning to the spatial vicinity of inputs and outputs for each task, effectively partitioning the network into weakly overlapping functional regions. Using simulations of tunable resistor networks, we demonstrate that this strategy enables robust memory of multiple sequential tasks while reducing the number of edges and the overall tuning cost. Our results hint at constrained training as a simple, local, and scalable mechanism to overcome catastrophic forgetting in tunable matter.
title Remembrance of Tasks Past in Tunable Physical Networks
topic Disordered Systems and Neural Networks
Soft Condensed Matter
Statistical Mechanics
url https://arxiv.org/abs/2512.03799