Hardware-In-The-Loop Training of a 4f Optical Correlator with Logarithmic Complexity Reduction for CNNs

This work evaluates a forward-only learning algorithm on MNIST with hardware-in-the-loop training of a 4f optical correlator, achieving 87.6% accuracy with O(n2) complexity, compared to backpropagation, which achieves 88.8% accuracy with O(n2 log n) complexity.

• Hardware-In-The-Loop Training of a 4f Optical Correlator with Logarithmic Complexity Reduction for CNNs
  L. Pes, M. Dehbashizadeh Chehreghan, R. Luiken, S. Stuijk, R. Stabile, and F. Corradi.
  In Optical Fiber Communications Conference, OFC 25 Proceedings, pages xyz-xyz. San Francisco, USA, 30 March - 3 April 2025. IEEE Computer Society Press, Los Alamitos, CA, USA, 2025. (abstract, pdf, doi).