Presentation Information
[TuP-G-5]Device-Informed Training Strategy for Low-Precision Photonic Neural Networks
Vikas Kumar Keshari1, Dhrutisundar Sahoo1, Abhishek Kumar Anand1, ○Lorenzo De Marinis2, Emilio Paolini2, Jimson Mathew1, Nicola Andriolli3, Giampiero Contestabile2, Sumanta Gupta1 (1.Indian Institute of Technology Patna, India, 2.Scuola Superiore Sant’Anna, Pisa, Italy, 3.University of Pisa, Pisa, Italy)
Keywords:
Photonics for neuromorphic computing and deep learning,Photonics in neuromorphic computing and machine learning devices
This paper compares three training strategies for low-precision photonic neural networks. The efficacy of the schemes is tested through circuital simulations on the photonic electronic multiply-accumulate neuron (PEMAN) architecture with three different datasets.