Extended Data Table 1 Deep ONNs

From: Fully forward mode training for optical neural networks

  1. a) Each optical layer performs 160,000-D to 160,000-D linear connection and optoelectronic nonlinear activations.
  2. b) Each optical layer nonlinearly transformed 100-D input to 50-D output, with linear transformations executed on electronic computer.
  3. [Science, 2021]62, [Nat. Photonics, 2022]63, [Nature, 2018]64.
Back to article page