Gradients of unitary optical neural networks using parameter-shift rule
This work addresses a bottleneck in optical computing by enabling efficient hardware-based training for researchers and engineers in photonics, though it is incremental as it adapts an existing technique to a specific domain.
The paper tackles the challenge of training unitary optical neural networks (UONNs) by adapting the parameter-shift rule (PSR) to compute exact analytical gradients directly from hardware measurements, offering an alternative to traditional methods like backpropagation that face physical constraints in optical systems.
This paper explores the application of the parameter-shift rule (PSR) for computing gradients in unitary optical neural networks (UONNs). While backpropagation has been fundamental to training conventional neural networks, its implementation in optical neural networks faces significant challenges due to the physical constraints of optical systems. We demonstrate how PSR, which calculates gradients by evaluating functions at shifted parameter values, can be effectively adapted for training UONNs constructed from Mach-Zehnder interferometer meshes. The method leverages the inherent Fourier series nature of optical interference in these systems to compute exact analytical gradients directly from hardware measurements. This approach offers a promising alternative to traditional in silico training methods and circumvents the limitations of both finite difference approximations and all-optical backpropagation implementations. We present the theoretical framework and practical methodology for applying PSR to optimize phase parameters in optical neural networks, potentially advancing the development of efficient hardware-based training strategies for optical computing systems.