Mediaspace scheduled maintenance: Aug 25, 2026 07:00 - 12:00 AM. During this time, videos will be temporarily unavailable. Check status updates.

Lecture

Diffractive Deep Network: Optical Computing

In course

MICRO-722: 3D Printing with light

Optical aspects of 3D printing technology. This includes optical systems for scanning and excitation, photopolymers, glass and other photoactive materials, and optical components fabricated with 3D printing technology. ...

Description

This lecture by the instructor covers the implementation of neural networks using diffractive layers and coherent light, focusing on optical computing. It explores the design of task-specific optical systems, all-optical machine learning, and diffractive optical neural networks. The presentation also includes experiments for classification, diffractive layers printing, and MNIST digit results.

Instructors (2)

Demetri Psaltis

Christophe Moser

Dr. Christophe Moser is Associate Professor of Microengineering and Industrial Relations at the Swiss Federal Institute of Technology in Lausanne. He was the co-founder and CEO of Ondax,Inc for ten years prior to joining EPFL. During this decade, he raised $15 million from corporate and venture capital sources to fund volume production of thick holographic components in glass and develop devices enabled by these components such as tunable add-drop multiplexers for telecommunications, ultra-narrowband notch filters for low frequency Raman systems, high power semi-conductor frequency-narrowed pumps for fiber lasers, single frequency fixed and tunable lasers for sensing and femtosecond pulse compressors for micro-machining applications. Dr. Moser has a Ph.D. in Electrical Engineering from the California Institute of Technology, a minor in finance and a bachelor degree in Physics from EPFL. He is the co-inventor of 24 patents, 14 peer reviewed scientific publications, 9 IEEE proceedings and a book chapter. Private interests include tennis, running, ski, basketball and family excursions. 2010: Associate Professor, EPFL, faculty Science Technique Ingenieur. 2000-2010: co-founder, CEO Ondax, Inc. 1995-2000: Graduate Research Assistant, California Institute of Technology. 1993-1995: Engineering Project manager , TESA. 1988-1993: Bachelor candidate in Physics, EPFL. MAIN PUBLICATIONS:

Moser C., Havermeyer F,.Ultra-narrowband tunable laserline notch filter, Appl. Phys. B, 95 (3), pp 597-601, 2009.

Moser C. ,Ho L., Havermeyer F,. Self-aligned Non-dispersive External Cavity Tunable Laser, 16 (21), 16691-16696 Optics Express, 2008.

Steckman, GJ , Moser C. et al. Volume holographic grating wavelength stabilized laser diodes, IEEE J. Of Selected Topics in Quantum Electronics, (13), 672-678, 2007.

Buse K, Havermeyer F, Liu W., Moser C, Psaltis D. Holographic Filters , Book Chapter Photorefractive Materials and their Applications, 2005.

Havermeyer, F; Liu, WH; Moser, C, et al. Volume holographic grating-based continuously tunable optical filter , Opt. Eng. 43 (9), 2017-2021, 2004.

Moser C., Maravic I., Schupp B., Adibi A, Psaltis D, Diffraction efficiency of localized holograms in doubly doped LiNbO3 crystals, Opt. Lett. 25: (17),1243-1245, 2000.

Official source

Ontological neighbourhood

Information engineering

Machine learning: Artificial neural networks

Related concepts (32)

Printing

Printing is a process for mass reproducing text and images using a master form or template. The earliest non-paper products involving printing include cylinder seals and objects such as the Cyrus Cylinder and the Cylinders of Nabonidus. The earliest known form of printing as applied to paper was woodblock printing, which appeared in China before 220 AD for cloth printing. However, it would not be applied to paper until the seventh century.

Deep learning

Deep learning is part of a broader family of machine learning methods, which is based on artificial neural networks with representation learning. The adjective "deep" in deep learning refers to the use of multiple layers in the network. Methods used can be either supervised, semi-supervised or unsupervised.

Handwriting recognition

Handwriting recognition (HWR), also known as handwritten text recognition (HTR), is the ability of a computer to receive and interpret intelligible handwritten input from sources such as paper documents, photographs, touch-screens and other devices. The image of the written text may be sensed "off line" from a piece of paper by optical scanning (optical character recognition) or intelligent word recognition. Alternatively, the movements of the pen tip may be sensed "on line", for example by a pen-based computer screen surface, a generally easier task as there are more clues available.

Recurrent neural network

A recurrent neural network (RNN) is one of the two broad types of artificial neural network, characterized by direction of the flow of information between its layers. In contrast to uni-directional feedforward neural network, it is a bi-directional artificial neural network, meaning that it allows the output from some nodes to affect subsequent input to the same nodes. Their ability to use internal state (memory) to process arbitrary sequences of inputs makes them applicable to tasks such as unsegmented, connected handwriting recognition or speech recognition.

Machine learning

Machine learning (ML) is an umbrella term for solving problems for which development of algorithms by human programmers would be cost-prohibitive, and instead the problems are solved by helping machines 'discover' their 'own' algorithms, without needing to be explicitly told what to do by any human-developed algorithms. Recently, generative artificial neural networks have been able to surpass results of many previous approaches.

Related lectures (32)

Neural Networks: Training and Activation

Explores neural networks, activation functions, backpropagation, and PyTorch implementation.

Neural Networks: Two Layers Neural Network

Covers the basics of neural networks, focusing on the development from two layers neural networks to deep neural networks.

Deep Learning: Convolutional Neural Networks

Covers Convolutional Neural Networks, standard architectures, training techniques, and adversarial examples in deep learning.

Neural Networks: Multilayer Learning

Covers the fundamentals of multilayer neural networks and deep learning, including back-propagation and network architectures like LeNet, AlexNet, and VGG-16.

Statistical Physics in Machine Learning: Understanding Deep Learning