Limits of nonlinear and dispersive fiber propagation for an optical fiber-based extreme learning machine
By: Andrei V. Ermolaev , Mathilde Hary , Lev Leybov and more
Potential Business Impact:
Computers learn to read numbers using light.
We report a generalized nonlinear Schr\"odinger equation simulation model of an extreme learning machine (ELM) based on optical fiber propagation. Using the MNIST handwritten digit dataset as a benchmark, we study how accuracy depends on propagation dynamics, as well as parameters governing spectral encoding, readout, and noise. For this dataset and with quantum noise limited input, test accuracies of : over 91% and 93% are found for propagation in the anomalous and normal dispersion regimes respectively. Our results also suggest that quantum noise on the input pulses introduces an intrinsic penalty to ELM performance.
Similar Papers
On-chip wave chaos for photonic extreme learning
Optics
Makes computers learn faster using light.
Optical Physics-Based Generative Models
Optics
Makes AI learn faster and use less power.
Training nonlinear optical neural networks with Scattering Backpropagation
Optics
Teaches light-based computers to learn faster.