Fault-Tolerant Deep Photonic Computing on a Single Physics-Informed Metasurface
Fault-Tolerant Deep Photonic Computing on a Single Physics-Informed Metasurface
This archive contains the reproducibility package for the research project: "Fault-Tolerant Deep Photonic Computing on a Single Physics-Informed Metasurface". The package includes: • Core photonic neural network implementations• Physics-Informed Neural Network (PINN) calibration models• SLiM (Single-Layer Integrated Metasurface) architecture• Experiment scripts and notebooks• Training results and benchmarking outputs• Figures used in the manuscript Key results reproduced by this codebase: • Recovery of photonic neural network accuracy from 41.73% to 95.27% using physics-informed calibration• Extinction ratio improvement to 63.96 dB• Implementation of a 100-layer virtual optical neural network using the SLiM architecture• Energy efficiency of approximately 16.2 fJ/MAC The code is implemented in Python using PyTorch and was tested on NVIDIA RTX GPUs. GitHub repository:https://github.com/raviraja1218/-photonic_slim
optical accelerators, photonic AI, photonic computing, physics informed neural networks, optical neural networks, photonic integrated circuits, metasurface computing, analog computing
optical accelerators, photonic AI, photonic computing, physics informed neural networks, optical neural networks, photonic integrated circuits, metasurface computing, analog computing
selected citations These citations are derived from selected sources.This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).0 popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.Average influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).Average impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.Average
Citations provided by BIP!Popularity provided by BIP!