In the context of the European training network PostDigital+, we are looking for several highly motivated PhD candidates with a background in photonics and an interest in machine learning or in combinatorial optimisation, for several research projects in the field of non-conventional photonic computing.

As a group, we have been active in this field for over 10 years, both at the simulation level and at the experimental level. E.g., we were the first to fabricate silicon photonics chips based on reservoir computing (a variant of recurrent neural networks) [Vandoorne et al., Nature Communications 5 3541, 2014]. The main application for this technology that we have been pursuing recently is telecommunication. Indeed, we used these neuromorphic chips to show the experimental realisation of nonlinear dispersion compensation in a telecom link [S. Sackesyn et al., Optics Express, 29, 20, 30991, 2021] and [E. Gooskens et al., Scientific Reports, 13, 21399, 2023]. Currently, we are closely collaborating with telecom companies to bring this closer to market. We are looking for highly motivated people to help us in that endeavour, so that we can offload costly processing by digital electronic DSP to analog photonic hardware.

A second strand of non-conventional photonic information processing is related to coherent Ising machines. These are based on the idea that the lowest energy state of a well-constructed dynamical system can encode the solution to a complex combinatorial optimisation problem. Such problems are very relevant for industry, e.g. for logistics, finance, biomedical research, … . However, finding a solution conventionally requires high-performance computer clusters that consume large amounts of energy and run for a long time. Using photonics, we want to speed up these computationally difficult tasks in a way unlike any current digital computer, i.e. by letting a hardware system with binary variables relax to the minimum energy state, at the same time finding the solution to the optimisation problem. However, more research is needed to achieve this, and we are currently e.g. investigating higher-order or multilevel Ising machines.  

In summary, we are looking for motivated and talented candidates to strengthen our team and to bring these research lines forward. Based on your skills and interest, we can match you with a current or new research line in our group.  We offer you the opportunity to perform cutting-edge, blue-sky research, in a challenging, motivating environment, working within a multidisciplinary team consisting of both photonics people and computer scientists. A willingness to tackle challenges coming from these multidisciplinary collaborations is a must.