Day: January 27, 2020

Faculty Candidate Fan Chen to Present Lecture February 10

The Colleges of Computing and Engineering invite the campus community to a lecture by faculty candidate Fan Chen, Monday, February 10, 2020, at 3:00 p.m., in Chem. Sci. 102. Chen’s talk is titled, “Efficient Hardware Acceleration of Unsupervised Deep Learning.”

Chen is a Ph.D. candidate in the Department of Electrical and Computer Engineering, Duke University, where she is advised by Professor Yiran Chen and Professor Hai “Helen” Li. Her research interests include computer architecture, emerging nonvolatile memory technologies, and hardware accelerators for machine learning. Fan won the Best Paper Award and the Ph.D. forum Best Poster Award at ASP-DAC 2018. She is a recipient of the 2019 Cadence Women in Technology Scholarship.

Abstract: Recent advances in deep learning are at the core of the latest revolution in various artificial intelligence (AI) applications including computer vision, autonomous systems, medicine, and other key aspects of human life. The current mainstream supervised learning relies heavily on the availability of labeled training data, which is often prohibitively expensive to collect and accessible to only a few industry giants. The unsupervised learning algorithm represented by Generative Adversarial Networks (GAN) is seen as an effective technique to obtain a learning representation from unlabeled data. However, the effective execution of GANs poses a major challenge to the underlying computing platform.

In her talk, Chen will discuss her work that devises a comprehensive full-stack solution for enabling GAN training in emerging resistive memory based main memory. A zero-free dataflow and pipelined/parallel training method is proposed to improve resource utilization and computation efficiency. Hao will also introduce an inference accelerator that enables developed deep learning models to run on edge devices with limited resources. Finally, Hao’s lecture will discuss her vision of incorporating hardware acceleration for emerging compact deep learning models, large-scale decentralized training models, and other application areas.


Faculty Candidate Cong “Callie” Hao to Present Lecture February 17

The Colleges of Computing and Engineering invite the campus community to a lecture by faculty candidate Cong (Callie) Hao, Monday, February 17, 2020, at 3:00 p.m., in Chem Sci 102. Hao’s talk is titled, “NAIS: neural architecture and implementation search.”

Dr. Hao is a postdoctoral researcher in the Department of Electrical and Computer Engineering (ECE) at University of Illinois at Urbana-Champaign (UIUC), under the supervision of Prof. Deming Chen. She holds a PhD (2017) degree in electrical engineering from Waseda University, and M.S. and B.S. degrees in computer science and engineering from Shanghai Jiao Tong University. Her research interests include high-performance reconfigurable computing, hardware-aware machine learning and acceleration, electronic design automation (EDA) tools, and autonomous driving.

Abstract: In her talk, Hao introduces a neural network and hardware implementation co-search methodology, named NAIS, to pursue aggregated solutions of high accuracy DNN designs and efficient hardware deployments simultaneously. To enable a comprehensive co-search framework, there are three indispensable components: 1) efficient hardware accelerator design (e.g. FPGA); 2) hardware-aware neural architecture search (NAS); and 3) automatic design tools to quickly deploy DNNs to hardware platforms. I will discuss each component and their integrations to support an efficient and optimal NAIS implementation.


Faculty Candidate Jean Hardy to Present Lecture Feb. 3

The Colleges of Computing and Engineering invite the campus community to a lecture by faculty candidate Jean Hardy, Monday, February 3, 2020, at 3:00 p.m., in EERC 214. Hardy’s talk is titled “What does community-driven technological development look like in rural Michigan?”

Hardy is PhD candidate and Rackham Merit Fellow in the University of Michigan (UM) School of Information. His research uses ethnographic and participatory design methods to understand how people use information and communication technologies for community formation and economic development in the rural Midwestern United States.

He is the co-organizer of UM’s Rural America Working Group and is affiliated with the UM Program in Science, Technology, and Society. Hardy’s formative work in rural computing has been published in Information, Communication, & Society and the Proceedings of the ACM on Human-Computer Interaction. He was awarded Best Paper Honorable Mention Awards at CHI 2016 and CSCW 2017, and a Best Provocation award at the 2019 ACM Conference on Designing Interactive Systems.

Lecture Abstract: The growth of the digital economy and the adoption of digital technologies continue to be widely regarded as opportunities to shift economic prospects, invigorate communities, and have transformative effects on society as we know it. But, digital technology, and the infrastructure that supports it, is largely designed and built for urban assumptions of scale, connectivity, and density. This presentation asks, what does digital technology and infrastructure look like when it leaves the urban and enters the rural, and how do we respond to the unique digital needs and aspirations of people living in rural communities?

Drawing from ethnographic and participatory design research in the Upper Peninsula of Michigan, I show how assumptions of growth and scalability built into contemporary social technologies do not reflect the reality of rural communities. I demonstrate how community-based research methods can be used to better understand the aspirations and needs of the people living in rural areas. I argue that the corporate obsession with scalability in contemporary social technologies misplaces opportunity for creative and unique digital tools that can engender a diverse future rural society.