paper-with-me

Papers

Training Large-Scale Optical Neural Networks with Two-Pass Forward Propagation

2024-08-15 · Amirreza Ahmadnejad, Somayyeh Koohi

This paper addresses the limitations in Optical Neural Networks (ONNs) related to training efficiency, nonlinear function implementation, and large input data processing. We introduce Two-Pass Forward Propagation, a novel training method that avoids specific nonlinear activation functions by modulating and re-entering error with random noise. Additionally, we propose a new way to implement convolutional neural networks using simple neural networks in integrated optical systems. Theoretical foundations and numerical results demonstrate significant improvements in training speed, energy efficiency, and scalability, advancing the potential of optical computing for complex data tasks.

📄 PDF Abstract BibTeX arXiv:2408.08337

Code (1)

AAhmadnejad98/Training-Large-Scale-ONN 공식 구현

Similar Papers 제목 키워드 기반

Large-scale nonlinear optical computing with incoherent light via linear diffractive systems

2026-03-31 · Alexander Chen, Yuntian Wang, Md Sadman Sakib Rahman, Yuhang Li 외 arxiv

Nonlinear computation is essential for various information processing tasks. Optical implementations are attractive because passive light propagation can manipulate high-dimensional signals with extreme throughput and pa…

Uncertainty Estimates and Multi-Hypotheses Networks for Optical Flow

2018-02-20 · ECCV 2018 9 · Eddy Ilg, Özgün Çiçek, Silvio Galesso, Aaron Klein 외

Optical flow estimation can be formulated as an end-to-end supervised learning problem, which yields estimates with a superior accuracy-runtime tradeoff compared to alternative methodology. In this paper, we make such ne…

Optical Flow Estimation

A silicon photonics feed-forward neural network for nonlinear distortion mitigation in an optical link

2020-12-10 · Mattia Mancinelli, Paolo Bettotti, Lorenzo Pavesi

We design and model a single-layer, passive, all-optical silicon photonics neural network to mitigate optical link nonlinearities. The network nodes are formed by silicon microring resonators whose transfer function has …

Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit

2020-08-26 · Tiankuang Zhou, Xing Lin, Jiamin Wu, Yitong Chen 외

Application-specific optical processors have been considered disruptive technologies for modern computing that can fundamentally accelerate the development of artificial intelligence (AI) by offering substantially improv…

ERASE: EaRly bAckpropagation SchEdule for Faster Training of Modern Recommendation Systems

2026-08-19 · Ergan Shang, Flavio Sales Truzzi arxiv

Lightweight proxy models enable rapid experimentation without repeatedly training frontier-scale systems, but their small kernels often leave modern accelerators underutilized. Conventional training compounds this ineffi…

Recommendation Systems