Deep Neural Networks have received much attention and
investment from the research community as well as industry in recent years
owing to the highly accurate performance in certain classes of machine
perception tasks. This coupled with the ever-increasing demand for smart
systems is driving the need for continuous improvement in performance
while requiring the technology to be cheaper and energy-efficient and thus
more widely available in portable/nomadic applications. A large part of these
advancements have relied on the improvement of computing hardware over the
last few decades and the majority of CNN/DNNs today run on high-performance
computing platforms, like multi-core CPUs and GPUs. However a push exists
towards reducing the cost and especially energy-efficiency.
This has given rise to a new and interesting research problems that require
pushing the boundaries of classical architecture design paradigms and to
co-optimize them together with circuit design and technology implementation.
co-optimisation direction. A pipelined data-flow scheme which eliminates
the need for costly local (SRAM) memory accesses during the tensor convolution
execution will be used as basis. But many ways exist to project that
data-flow into a cost/energy-effective architecture and circuit.
We want to explore that broad search space for the context of (3+1)-D
convolutions. We also want to exploit emerging technology options which
support effective use of the 3rd scaling dimension. That will
enable potentially strong gains in the interconnections which typically
dominate the realisation of large processing networks like CNN/DNNs.
Required background: electrical or microelectronic engineer with a strong background in the domain of processor architecture and circuit design concepts and strong interest in novel fabrication technologies.
Type of work: 30% architecture design, 30% circuit design, 30% tape out/measurements, 10% literature
Supervisor: Francky Catthoor
Daily advisor: Stefan Cosemans
The reference code for this position is 1812-87. Mention this reference code on your application form.