PulseAugur
LIVE 07:20:30
research · [1 source] ·
0
research

ROSA optical neural network architecture boosts efficiency and robustness

Researchers have introduced ROSA, a novel microring-based optical neural network architecture designed for enhanced robustness and energy efficiency. This design incorporates an optical shift-and-add module and a layer-wise hybrid mapping strategy. The system optimizes performance by considering noise-aware voltage-to-weight modeling and co-optimizing array size with dataflow, leading to significant reductions in energy-delay product and improvements in accuracy on benchmarks like CIFAR-10. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Introduces a new hardware architecture for neural networks that significantly improves energy efficiency and accuracy.

RANK_REASON This is a research paper detailing a new architecture for optical neural networks.

Read on arXiv cs.LG →

COVERAGE [1]

  1. arXiv cs.LG TIER_1 · Huifan Zhang, Yun Hu, Caizhi Sheng, Yurui Qu, Pingqiang Zhou ·

    ROSA: Robust and Energy-Efficient Microring-Based Optical Neural Networks via Optical Shift-and-Add and Layer-Wise Hybrid Mapping

    arXiv:2605.00032v1 Announce Type: cross Abstract: This work presents ROSA, a microring-based optical neural network architecture that improves robustness and energy efficiency using an optical shift-and-add (OSA) module and a layer-wise hybrid mapping strategy. It introduces a no…