Computation by Photodetection: Deep Learning and Optimization with Coherent Photonic Signals
- Datum: 12.09.2022
- Uhrzeit: 11:00
- Vortragender: Dr. Ryan Hamerly
- NTT Research & Massachusetts Institute of Technology
- Ort: Max-Planck-Institut für Mikrostrukturphysik, Weinberg 2, 06120 Halle (Saale)
- Raum: Lounge
We propose a novel scheme for optical neural-network acceleration based on balanced homodyne detection, analyze its performance on speed and energy-efficiency benchmarks, and outline the prospects for realization in integrated photonics.
Dr. Ryan Hamerly is a senior scientist at NTT Research and a visiting scientist at MIT's Quantum Photonics Laboratory. He received the B.S. from Caltech in 2010 and Ph.D. in Applied Physics from Stanford In 2016. In 2017, he was at the National Institute of Informatics (Tokyo), working on quantum annealing and optical computing concepts. He completed a postdoctoral fellow at MIT with Prof. Dirk Englund before joining NTT Research. His current research interests are photonic AI accelerators and related topics.