PhD student on Ultra-low Power Deep Learning accelerators in mobile platforms (V36.3469)

PhD student on Ultra-low Power Deep Learning accelerators in mobile platforms

Aanstellingsfunctie(s)
Promovendus
Faculteit(en)
Faculteit Electrical Engineering
FTE
1,0
Sluitdatum
25/11/2018
Aanvraagnummer
V36.3469

Functieomschrijving

Eindhoven University of Technology has a vacancy for a

PhD student on Ultra-low Power Deep Learning accelerators in mobile platforms

in the Electronic Systems group, department of Electrical Engineering

Project description
There is still an enormous difference in energy-efficiency between modern deep learning (DL) ASICs and the human brain. Modern DL HW platforms cannot get close even when flexibility is sacrificed in their hardware architecture in favor of energy-efficiency. Furthermore, the high compute, storage, and memory bandwidth requirements of modern DL applications put a lot of stress on the energy budget and result in a significant hardware cost. Within the efficient Deep Learning Platforms (eDLP) project we aim to adopt principles that make the human brain so energy-efficient (i.e., low frequency, massively parallel, analog, asynchronous, approximate/low precision computing, redundancy/fault tolerance) in the design of HW/SW platforms for DL applications. Our goal is to achieve a huge improvement in power-efficiency of DL solutions, such that they become a viable option for power and cost restricted embedded systems.

Within the larger eDLP project, this PhD position will contribute to the implementation of ultra-low-power microcontroller platforms with embedded hardware accelerators for deep learning. The relative slow innovation progress on battery technologies demands radical innovations for energy-efficient and autonomous operation. On the semiconductor side of the solution (which is the focus of the eDLP project), these include a charge recycling strategy that can be used for example in combination with a trickle charger of a solid-state battery. Additional large gains will be obtained by applying approximate computing. Topics of research will include: charge recycling strategies based on stacked logic to reduce battery drainage, context aware approximate computing solutions for neural network algorithms that enable a trade-off between energy consumption and quality-of-experience, and event-driven HW accelerators that minimize the overall energy consumption.

Departments and collaborators
Eindhoven University of Technology (TU/e) is a world-leading research university specializing in engineering science & technology. The Department of Electrical Engineering is responsible for research and education in Electrical Engineering. The discipline covers technologies and electrical phenomena involved in computer engineering, information processing, energy transfer and telecommunication. The department strives for societal relevance through an emphasis on the fields of smart sustainable systems, the connected world and care & cure. The TU/e is the world’s best-performing research university in terms of research cooperation with industry (#1 since 2009).

The Electronic Systems group consists of seven full professors, ten assistant professors, several postdocs, about 50 PDEng and PhD candidates and support staff. The ES group is world-renowned for its design automation and embedded systems research. It is our ambition to provide a scientific basis for design trajectories of electronic systems, ranging from digital circuits to cyber-physical systems. The trajectories are constructive and lead to high quality, cost-effective systems with predictable properties (functionality, timing, reliability, power dissipation, and cost). Design trajectories for applications that have strict real-time requirements and stringent power constraints are an explicit focus point of the group. Within this area, prof.dr. H. Corporaal, prof.dr. J. Pineda and dr.ir. S. Stuijk have developed various novel power efficient computer architectures and their associated compilation trajectories.

The eDLP project team is designed to combine extensive knowledge in the key fields. The project team includes industrial partners from Holst, NXP, Intel, and NVIDIA. As part of this project, the PhD candidate will work closely with these industrial partners to ensure that the developed architecture is suitable for the systems developed by these partners.

Functie-eisen

We are looking for candidates that match the following profile:

  • A master's degree in Electrical Engineering or related disciplines with excellent grades.
  • Excellent knowledge of computer architectures and hardware design.
  • Solid programming skills (e.g., in C or C++).
  • Experience with hardware design (e.g., in Verilog or HLS).
  • Experience with ASIC design, e.g. logic synthesis, and place and route.
  • Transistor-level knowledge for circuit-level design, e.g. standard cells, energy monitors, and power conversion.
  • A team player that enjoys to work in multicultural teams.
  • Good communication and organization skills.
  • Excellent English language skills (writing and presenting).

Arbeidsvoorwaarden

We offer a fixed-term, 4 year position in a research group with an excellent reputation. Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities, including:

  • A gross monthly salary between EUR 2266,- (first year) and EUR 2897- (last year).
  • Additionally, 8% holiday and 8.3% end-of-year annual supplements.
  • A minimum of 41 holidays per year (excluding bank holidays, for a full-time employment of 40 hrs/week)
  • Additional benefits, including excellent technical infrastructure, child care, holiday savings schemes, and sports facilities.
  • Assistance for finding accommodation is offered.
  • Personal development program aimed to develop your social and communication skills (see http://www.tue.nl/PROOF3TU).

Informatie en sollicitatie

Information 

Application

If interested, please use 'apply now'-button at the top of this page. You should upload the following:

  • a detailed curriculum vitae, a letter of motivation and portfolio with relevant work;
  • a cover letter explaining your motivation and suitability for the position;
  • a detailed Curriculum Vitae (including a list of publications and key achievements in research project(s));
  • contact information of two references;
  • copies of diplomas with course grades;

Candidates will be selected based on graduation mark and proficiency at university including consideration of the reputation of the university, relevant experience and skills, writing skills and publications, work experience as well as performance in relevant modeling exercises and interviews.
 

Please keep in mind; you can upload only 5 documents up to 2 MB each.