2023/11/01 by Sam Dillavou, Benjamin D. Beyer, Benjamin D Beyer +7 · 1 voice · 5 citations
Computer Science · Engineering · Neuroscience · Physics and Astronomy · #Advanced Memory and Neural Computing #Neural Networks and Reservoir Computing #Neural dynamics and brain function #cond-mat.soft #cs.ET #cs.LG
paper · pdf · doi:10.1073/pnas.2319718121
arxiv published 2023/11/01 · arxiv updated 2024/04/05 · openalex publication_date 2024/07/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
Standard deep learning algorithms require differentiating large nonlinear networks, a process that is slow and power-hungry. Electronic contrastive local learning networks (CLLNs) offer potentially fast, efficient, and fault-tolerant hardware for analog machine learning, but existing implementations are linear, severely limiting their capabilities. These systems differ significantly from artificial neural networks as well as the brain, so the feasibility and utility of incorporating nonlinear elements have not been explored. Here, we introduce a nonlinear CLLN—an analog electronic network made of self-adjusting nonlinear resistive elements based on transistors. We demonstrate that the system learns tasks unachievable in linear systems, including XOR (exclusive or) and nonlinear regression, without a computer. We find our decentralized system reduces modes of training error in order (mean, slope, curvature), similar to spectral bias in artificial neural networks. The circuitry is robust to damage, retrainable in seconds, and performs learned tasks in microseconds while dissipating only picojoules of energy across each transistor. This suggests enormous potential for fast, low-power computing in edge systems like sensors, robotic controllers, and medical devices, as well as manufacturability at scale for performing and studying emergent learning.