About this course
After substantial photonic technology advances and the disruptive advent of artificial intelligence, the ever-desired ultra-fast ultra-wide band signal processing via neuromorphic optical neural networks is perceived as a reality. Knowing the optical processor architectures will make the students aware of their potential and to come up with novel and fully integrated systems/applications for disruptively low cost, low power consumption and ultra-fast computing.
The course is thought to teach the basis of the optical computation and of the brain-inspired optical computation and neuromorphic (quantum) photonics, drive the students along the progress of optical computation and make them aware of the huge advantages that this can offer. Use cases offered by companies in optical computation and artificial intelligence will help the students to think through the overall (electro-optical) system and come up with innovative system implementation proposals.
Moreover, the Dutch photonic eco-system includes a myriad of companies which are more and more interested in adopting optics and photonics as key enabling technology and for data processing. Having said so, this course is expected to have a double goal: to be of complement to disciplines within the new master AI, as well as to connect the students to the prolific optics/photonics Dutch eco-system.
Learning outcomes
General learning objectives:
The basis of optical computation and of brain-inspired computation in photonics and quantum photonicswill drive the students along the progress of photonics and integrated photonics in computation, making aware of the fact that ultra-fast ultra-wide band signal processing via neuromorphic optical neural networks is a reality.
Learning objective
The student will be able to:1. understand what is optical computation and in what sense it outperforms electrical computation;
2. understand optical computing using micro-optics (Fourier optics), as well as using photonic integrated chips and the performance gains of each of those;
3. learn new ways to do optical computations (via optical fibers, and quantum optical neural networks);
4. understand algebraic optical operations and different applications, as well as interfaces and constraints for the processing engine;
5. understand learning methods;
6. come up with a novel application which needs an optical computing engine and design the fully integrated system, as well as identify performance gains.
Resources
- Slides and published papers on IEEE or OSA journal papers. These same papers will be discussed to provide feedbacks for the group projects.