A hybrid diffractive-holographic optical architecture embeds learned phase patterns into passive elements to perform MNIST image classification at 91.2% accuracy via physical wavefront propagation with nanosecond latency.
All-optical machine learning using diffractive deep neural networks.Science, 361(6406):1004–1008
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Photonic AI: A Hybrid Diffractive Holographic Neural System for Passive Optical Real-Time Image Classification
A hybrid diffractive-holographic optical architecture embeds learned phase patterns into passive elements to perform MNIST image classification at 91.2% accuracy via physical wavefront propagation with nanosecond latency.