IEEE Photonics Society

Boston Photonics Society Chapter

Boston Chapter of the IEEE Photonics Society

Machine Learning and Optical Systems PDF

Wednesday, October 7, 14, 21, 28 and November 4, 2020, 7:00-9:30 PM
Located at Online Seminar

This workshop will feature talks on different aspects of the interaction between Machine learning (ML) and optical systems. This interplay, between how ML is used to improve optical systems and how optical systems are used to implement the deep neural network hardware (DNN) for ML, is one of the driving forces advancing optics applications.

ML including DNN provide a new set of tools to the photonics and optical communication community, which offer opportunities when the system is highly complex, or when there is a lack of analytical models. Recent applications of ML to optical materials, optical design, plasmonics, metasurface optics, optical communication systems, and optical measurements will be covered and will offer insights as to how various optical problems can benefit from ML.

Photons are ideal information carriers in distributed processors such as DNNs and quantum computers. There are a variety of optical approaches to accelerate DNNs and quantum computers, and significant effort has been made towards prototyping such systems. Some implementations offer massive parallelism, while others offer flexibility in a compact size. Examining these differences will show us how photonics could play a role in future computing.

This workshop will bring together leading experts to discuss the latest research in the intersecting fields of ML, DNN, photonics, and optical communications. It also aims to foster communication and collaboration through networking among the individual engineers and researchers attending. Learn more about the rapid advances in the application of: ML to photonics and optical communications; and the optical implementation of neural network computing, directly from the foremost researchers in the different specialties involved, by registering for and attending this workshop.

Wednesday
October 7, 2020

Optimized Photonics

Prof. Jelena Vučković, Stanford University, Stanford, CA
 

Mapping AI Workloads to a Photonic Matrix Multiplier

Dr. Gilbert Hendry, Lightelligence, Boston, MA
 
 
 

Wednesday
October 14, 2020

Deep Elastic Strain Engineering of Band Structure through Machine Learning

Prof. Ju Li, Massachusetts Institute of Technology, Cambridge, MA
 

Interfacing Photonics with Artificial Intelligence: A New Design Strategy for Photonic Metamaterials based on Deep Learning

Prof. Yongmin Liu, Northeastern University, Boston, MA
 
 
 

Wednesday
October 21, 2020

Automatic Design of Optoelectronic Materials with Atomistic Simulations and Deep Learning

Prof. Rafael Gomez-Bombarelli, Massachusetts Institute of Technology, Cambridge, MA
 

Photonic Accelerators for Machine Intelligence

Prof. Dirk Englund, Massachusetts Institute of Technology, Cambridge, MA
 

Nanophotonic Accelerators for Recurrent Ising Machines

Ms. Mihika Prabhu, Massachusetts Institute of Technology, Cambridge, MA
 
 
 

Wednesday
October 28, 2020

Physics-Based Machine Learning for Fiber-Optic Communication Systems

Dr. Christian Häger, Duke University, Durham, NC
 

Deep Learning-Based Approaches for Meta-Atom and Meta-Device Design Slides

Prof. Hualiang Zhang, University of Massachusetts, Lowell, MA
 
 
 

Wednesday
November 4, 2020

Advancing Optical Communication and Measurement Systems Using Machine Learning

Prof. Darko Zibar, Technical University of Denmark, Kongens Lyngby, Denmark
 

Deep Learning of Ultrafast Pulses with a Multimode Fiber

Prof. Hui Cao, Yale University, New Haven, CT

The workshop expenses have
been generously supported by:

 

MERL


For more information on the technical content of the workshop, contact either:
1) Keisuke Kojima, (kojima@merl.com), Chair
2) Ajay Garg, (ajay.sinclair.garg@ieee.org), Co-Chair
3) Dean Tsang, (tsang@ieee.org), Co-Chair
4) Bill Nelson, (w.nelson@ieee.org), Co-Chair