STAR: Astrocyte-Inspired State-Augmented Repair for Supervised Memristive AI Hardware Systems
Emerging Technologies
Summary
The authors study how memristive crossbar arrays, used for on-chip learning, develop permanent faults that damage stored information and reduce performance, which normal retraining cannot fix. Inspired by brain cells called astrocytes that help repair synaptic faults, they explore how these cells modulate neural activity to recover function. From this, they design a new repair method called STAR that nudges the surviving parts of the network to restore its previous behavior without needing to know which parts are broken. Their approach works with a local learning rule and can be applied to complex networks like convolutional ones.
Authors
Yusuf Ahmed Khan, Zhuangyu Han, Abhronil Sengupta
Abstract
Memristive crossbar arrays have emerged as a promising platform for efficient on-chip learning, enabling local learning rules such as Equilibrium Propagation (EP) to be realized without the memory overhead of conventional backpropagation. However, as these devices age, permanent stuck-at (SA) faults accumulate at individual synaptic elements, irreversibly corrupting the stored weights and degrading model performance in ways that in-situ retraining alone cannot address. Existing mitigation strategies either rely on hardware redundancy at significant area and power cost, or require explicit fault localization that is impractical to perform continuously on-chip. This paper adopts a brain-inspired fault recovery route motivated by self-repair functionalities enabled by astrocytes -- a type of glial cell. We couple a computational neuroscience study of astrocytic neuromodulation under permanent synaptic faults with an algorithmic recovery mechanism for faulted crossbar-mapped networks. Our computational neuroscience study characterizes how astrocytes modulate surviving synapses in a bidirectional EP-specific network setting under both SA-0 and high-conductance stuck-at fault conditions, revealing that recovery operates at the neural population level. Motivated by this observation, we propose a retraining-based repair mechanism that augments EP with an additional repair nudge anchored to pre-fault activation targets of the healthy network, encouraging surviving weights to collectively reconstruct pre-fault internal representations without any information of which weights are faulty. To our knowledge, STAR is the first astrocyte-inspired repair mechanism to operate under a supervised local learning rule and to scale to convolutional networks.