ROSA achieves up to 64% lower energy-delay product and 8.3% higher CIFAR-10 accuracy in microring optical neural networks by combining optical shift-and-add with workload-aware hybrid mapping.
Toward lifelong-sustainable electronic- photonic ai systems via extreme efficiency, reconfigurability, and ro- bustness
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
-
ROSA: Robust and Energy-Efficient Microring-Based Optical Neural Networks via Optical Shift-and-Add and Layer-Wise Hybrid Mapping
ROSA achieves up to 64% lower energy-delay product and 8.3% higher CIFAR-10 accuracy in microring optical neural networks by combining optical shift-and-add with workload-aware hybrid mapping.