ROSA: Robust and Energy-Efficient Microring-Based Optical Neural Networks via Optical Shift-and-Add and Layer-Wise Hybrid Mapping
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 noise-aware voltage-to-weight model considering DAC and thermal variations, and a workload-aware framework to co-optimize MRR array size and layer-wise dataflow. Optimized arrays reduce the aggregated relative energy-delay product (EDP) by 64% and 26% compared with DEAP-CNNs and a general compact array, respectively. OSA further contributes 29% EDP reduction. The proposed hybrid mapping strategy improves CIFAR-10 accuracy by 8.3% over weight-stationary mapping while achieving an average 54.7% lower EDP than DEAP-CNNs.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
A Comparative Analysis of Microrings Based Incoherent Photonic GEMM Accelerators
Several microring resonator (MRR) based analog photonic architectures have been proposed to accelerate general matrix-matrix multiplications (GEMMs) in deep neural networks with exceptional throughput and energy efficien…
Thermal Crosstalk Modelling and Compensation Methods for Programmable Photonic Integrated Circuits
Photonic integrated circuits play an important role in the field of optical computing, promising faster and more energy-efficient operations compared to their digital counterparts. This advantage stems from the inherent …
A silicon photonics feed-forward neural network for nonlinear distortion mitigation in an optical link
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 …
C2PO: Coherent Co-packaged Optics using offset-QAM-16 for Beyond PAM-4 Optical I/O
Co-packaged optics (CPO) has emerged as a promising solution for achieving the ultra-high bandwidths, shoreline densities, and energy efficiencies required by future GPUs and network switches for AI. Microring modulators…
Self-calibrated Microring Weight Function for Neuromorphic Optical Computing
This paper presents a microring resonator-based weight function for neuromorphic photonic applications achieving a record-high precision of 11.3 bits and accuracy of 9.3 bits for 2 Gbps input optical signals. The system …