Postdoc in heterogeneous computing for deep learning at MDH/Sweden

Position description:

Malardalen University is looking for a postdoc in the field of heterogeneous computing for deep learning in automation and robotics. The position will be within the research project "DeepMaker - Deep Learning Accelerator on Commercial Programmable Devices". DeepMaker aims to provide a framework which generates synthesizable accelerators of Deep Neural Networks (DNNs) that can be used for different FPGA fabrics. DeepMaker enables optimization and effective use of DNN acceleration in commercially available devices that can accelerate a wide range of applications without the need for costly FPGA reconfigurations. This will be accomplished through close collaboration and interaction between two research groups at MDH and 3 participating industrial companies. DeepMaker is hosted at MDH in the Embedded Systems (ES) research environment. ES is an internationally leading environment in embedded-systems research and host 14 research groups with over 100 researchers, focusing of various aspects of development of embedded systems.

As a postdoc, you will spend a minimum of 80% of your time on research. The rest will be spent on educational and/or administrative duties. The temporary employment is valid for 1 year with possible extension to additional year. This position may lead to a longer period and possibly to assistant professor if s/he integrates well to the group.

Qualifications requirements:

• The applicant is required to have a PhD degree in relevant fields.
• A successful applicant should have demonstrated practical skills of formulating and developing partitioned problems on heterogeneous architectures (like UltraScale or Zynq).
• Good knowledge in programming languages and hardware description language.
• Applicants need to be highly motivated and fluent in English, both written and spoken.
• Decisive importance is attached to personal suitability. We value the qualities that an even distribution of age and gender, as well as ethnic and cultural diversity, can contribute to the organisation.

Merit: Documented experience with neural network and multi-objective optimization is of high merit.

Application: Application is made online, follow the instruction:
http://www.mdh.se/hogskolan/jobb/lediga-jobb-1.103104?l=sv_SE&rmpage=job&rmjob=263

Deadline: 2018-04-30

Contact person:
Masoud Daneshtalab
Associate Professor
masoud.daneshtalab@mdh.se
http://www.idt.mdh.se/~md/