Learning, neural networks and optics
Professor Demetri Psaltis, Ecole Polytechnique Fédérale de Lausanne, Switzerland
Leuchs-Russell Auditorium, A.1.500, Staudtstr. 2
An optical imaging system generally measures a property of an object as a function of its special coordinates and makes this information available for further use. Conventionally, the design of the system relies on our understanding of the physical properties of the optical system that transfers the light scattered by the object to the detector. For example, since a lens undoes the blurring that occurs when a light field propagates through free space it is used to focus a distant object onto a 2D sensor directly forming an image. In other cases the situation is more complex and it is not possible to present to the detector directly a clear image of the unknown object. In such cases the detected signal must be processed further in order to extract the image. In this presentation we describe how we can perform imaging in such complex systems through the presentation of examples. We describe two cases. First we consider a multi-mode fiber as an imaging element and we show that we can learn to transmit or interpret arbitrary images sent through the fiber by training the fiber with a set of basis functions. Secondly we show that we can learn the shape of an object from examples formed by reconfiguring the optical system in a predictable way. We demonstrate this second modality by constructing a neural network that models the optical system and training the network to reproduce the experimentally measured data. The adaptable parameters of the trained network yield the image of the unknown object.
Demetri Psaltis is Professor of Optics and the Director of the Optics Laboratory at the Ecole Polytechnique Fédérale de Lausanne (EPFL). He was educated at Carnegie-Mellon University where he received the Bachelor of Science in Electrical Engineering and Economics in 1974, the Master’s in 1975, and the PhD in Electrical Engineering in 1977. In 1980, he joined the faculty at the California Institute of Technology, in Pasadena California where he held the Thomas G. Myers Professor of Electrical Engineering. He served as the Executive Officer for the Computation and Neural Systems department from 1992–1996. From 1996 until 1999 he was the Director of the National Science Foundation research center on Neuromorphic Systems Engineering at Caltech. At Caltech in 2004 he established the Center for Optofluidic Integration and he served as the director until he moved to EPFL in 2006 where he established his research lab and served as Dean of the engineering school for ten years. His research interests are in imaging, holography, biophotonics, nonlinear optics and optofluidics. He has over 400 publications in these areas. Dr. Psaltis is a fellow of the IEEE, the Optical Society of America, the European Optical Society and the Society for Photo-Optical Systems Engineering (SPIE). He received the International Commission of Optics Prize, the Humboldt Award, the Leith Medal, th e Gabor Pri ze and the Joseph Fraunhofer Award / Robert M. B urley Prize.
MEETING THE SPEAKER
The lectures will follow a colloquium format for a broad audience and will be followed by a reception to provide an opportunity for meeting the speaker.