The aim of this PhD studentship is to develop systems capable of operating on probability distributions much like current systems operate on scalars and vectors. These systems will be used for computing directly in the probability domain. Such systems could enable modern electronics to deal with ambiguity much more naturally, as well as much more reliably than current approaches. We anticipate this to become a key next-generation AI hardware accelerator type and has the potential to be applied in a variety of environments, especially safety-critical ones ranging from autonomous vehicles to industrial robotics and medical devices. The project covers a wide spectrum of experimental research, including mixed-signal IC and/or embedded design, machine learning and programming. The PhD student will have the opportunity to join a multi-disciplinary team and to be trained and work in the world-class facilities of the Zepler Institute for Photonics and Nanoelectronics.
We welcome applications from candidates with a background in electronics, computer science, physics and maths. Prior experience with programming (Python, Tensor Flow), circuit design and data science are highly desired. Interested candidates are encouraged to contact Prof Themis Prodromakis (firstname.lastname@example.org) or Dr Alex Serb (email@example.com).