Low-rank surrogate modeling and stochastic zero-order optimization for training of neural networks with black-box layers

ArXi:2509.15113v2 Announce Type: replace The growing demand for energy-efficient, high-performance AI systems has led to increased attention on alternative computing platforms (e.g., photonic, neuromorphic) due to their potential to accelerate learning and inference. However, integrating such physical components into deep learning pipelines remains challenging, as physical devices often offer limited expressiveness, and their non-differentiable nature renders on-device backpropagation difficult or infeasible.