Noise-Aware Training of Neuromorphic Dynamic Device Networks
This work addresses the problem of programming dynamical device networks for researchers in physical computing, offering an incremental improvement by decoupling device model training from network training.
The paper tackled the challenge of designing networks of interconnected physical devices for dynamic tasks without accurate models or noise quantification, by proposing a noise-aware training methodology using Neural Stochastic Differential Equations as differentiable digital twins, which reduced required training data and provided a robust framework validated on spintronic device networks across temporal benchmarks.
Physical computing has the potential to enable widespread embodied intelligence by leveraging the intrinsic dynamics of complex systems for efficient sensing, processing, and interaction. While individual devices provide basic data processing capabilities, networks of interconnected devices can perform more complex and varied tasks. However, designing networks to perform dynamic tasks is challenging without physical models and accurate quantification of device noise. We propose a novel, noise-aware methodology for training device networks using Neural Stochastic Differential Equations (Neural-SDEs) as differentiable digital twins, accurately capturing the dynamics and associated stochasticity of devices with intrinsic memory. Our approach employs backpropagation through time and cascade learning, allowing networks to effectively exploit the temporal properties of physical devices. We validate our method on diverse networks of spintronic devices across temporal classification and regression benchmarks. By decoupling the training of individual device models from network training, our method reduces the required training data and provides a robust framework for programming dynamical devices without relying on analytical descriptions of their dynamics.