Photonic neural networks offer energy-efficiency but suffer from noise-induced low precision. We propose AQ-PANN, which learns a quantization step size to mitigate noise. Experiments on SVHN show strong performance across bitwidths under different noise levels.
Mitigating Noise Effects in Photonic Neural Networks Using Adaptive Quantization
Andriolli N.Ultimo
2025-01-01
Abstract
Photonic neural networks offer energy-efficiency but suffer from noise-induced low precision. We propose AQ-PANN, which learns a quantization step size to mitigate noise. Experiments on SVHN show strong performance across bitwidths under different noise levels.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
