Postdoctoral Fellow in Accelerated Deep Learning at the National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign.
Deadline for application December 31st 2019.
The National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign provides supercomputing and advanced digital resources for the nation's scientific enterprise. At NCSA, University of Illinois faculty, staff, students, and collaborators from around the globe use advanced digital resources to address and research grand challenges for the benefit of science and society. NCSA has been advancing one third of Fortune 50® companies for more than 30 years by bringing industry, researchers and students together to solve grand challenges at rapid speed and scale.
NCSA is looking for a talented and motivated postdoctoral fellow to lead a research and development effort in the area of accelerated deep learning. This position will contribute to the development and implementation of deep learning algorithms, and their optimization on GPUs and FPGAs. GPU- and FPGA-based low-latency inference machine learning algorithms will be developed in support of a newly funded effort to enable new capabilities in big-data physics experiments. The successful candidate is expected to lead a vibrant research program, propose major innovations, collaborate with peers within the group and across institutions participating in the project, publish results in top tier conferences, and contribute to proposals.
The selected candidate will become a member of the Center for Artificial Intelligence at NCSA, and will interact closely with the NCSA Gravity Group, pioneers in the use of AI and HPC for gravitational wave astrophysics, the Innovative Systems Lab, the central hub of innovation on emergent hardware architectures for AI and HPC at NCSA, and faculty from the Departments of Computer Science, Electrical and Computer Engineering, and Physics at Illinois. As part of this position, the postdoc will interact closely with a network of collaborators at MIT and the University of Washington, and will be encouraged to participate in cross disciplinary research with the NCSA Industry Program, the largest HPC industrial program in the world.
The successful candidate will work at NCSA with a main focus on the development and implementation of machine/deep learning algorithms for GPUs and FPGAs. He/she will be primarily engaged in the following activities:
Conduct research to push back the frontiers of machine learning for big-data science experiments
Adapt and innovate machine learning algorithms and frameworks on acceleration hardware for low-latency high throughput inference
Work with science teams to gather application requirements, and develop solutions that meet such metrics
Publish research results in high profile journals and conferences
The appointment will be for a fixed term of two years with the possibility of extension for another year. The salary range will be commensurate with the experience and qualifications of the candidate.
Required education and experience
PhD degree in computer science, computer engineering, computational sciences, or related fields
Experience developing GPU-based solutions using CUDA, OpenCL or OpenACC
Experience developing FPGA-based solutions using hardware design (Verilog, HDL) or HLS (OpenCL, Vivado HLS) methodology
Experience in C/C++/Python programming
Experience in deep neural networks (CNN, RNN), deep learning methodologies ((un)supervised learning, reinforcement learning, network reduction), and deep neural network optimization and acceleration
Analytical problem solving ability
Strong verbal and written communication skills
Good interpersonal and teamwork skills
Experience in high performance computing
Experience in the development of data-intensive scientific applications
Experience with Caffe, TensorFlow, or other machine learning frameworks
For full consideration, interested applicants should send their application materials (cover letter, curriculum vitae, research statement, links to recent publications, names/contact information of three referees) to postdoc-search(AT)ncsa.illinois.edu and include "NCSA Postdoc Application in Deep Learning" in the subject line. Interviews and hires may take place prior to the closing date and early applications are encouraged. Full consideration will be given to complete applications received by the closing date, December 31, 2019. For further information regarding position requirements and application procedures, contact Volodymyr Kindratenko at kindrtnk(AT)illinois.edu or +1 (217) 265-0209.
Please find here full details: http://www.ncsa.illinois.edu/about/jobs/pd_gravity