Scalable Back-Propagation-Free Training of Optical Physics-Informed Neural Networks
This work addresses the problem of enabling energy-efficient, real-time PINN training on edge devices for applications in solving PDEs, representing a novel method for a known bottleneck in photonic computing.
The paper tackled the challenge of training physics-informed neural networks (PINNs) on photonic chips by proposing a back-propagation-free framework, achieving real-time training and significant chip area reduction through innovations like a sparse-grid Stein derivative estimator and photonic tensor cores.
Physics-informed neural networks (PINNs) have shown promise in solving partial differential equations (PDEs), with growing interest in their energy-efficient, real-time training on edge devices. Photonic computing offers a potential solution to achieve this goal because of its ultra-high operation speed. However, the lack of photonic memory and the large device sizes prevent training real-size PINNs on photonic chips. This paper proposes a completely back-propagation-free (BP-free) and highly salable framework for training real-size PINNs on silicon photonic platforms. Our approach involves three key innovations: (1) a sparse-grid Stein derivative estimator to avoid the BP in the loss evaluation of a PINN, (2) a dimension-reduced zeroth-order optimization via tensor-train decomposition to achieve better scalability and convergence in BP-free training, and (3) a scalable on-chip photonic PINN training accelerator design using photonic tensor cores. We validate our numerical methods on both low- and high-dimensional PDE benchmarks. Through circuit simulation based on real device parameters, we further demonstrate the significant performance benefit (e.g., real-time training, huge chip area reduction) of our photonic accelerator.