Hardware-in-the-loop training of a 4f optical correlator with the PEPITA forward-only algorithm reaches MNIST accuracy comparable to backpropagation while removing the need for software gradients and claiming a log-factor reduction in training complexity.
Suitability of forward-forward and pepita learning to mlcommons-tiny benchmarks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Hardware-In-The-Loop Training of a 4f Optical Correlator with Logarithmic Complexity Reduction for CNNs
Hardware-in-the-loop training of a 4f optical correlator with the PEPITA forward-only algorithm reaches MNIST accuracy comparable to backpropagation while removing the need for software gradients and claiming a log-factor reduction in training complexity.