Events for the 4th week of March
-
ECE-EP Seminar - Volker Sorger, Monday, March 21st @ 2pm in EEB 248
Mon, Mar 21, 2022 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Volker Sorger, George Washington University
Talk Title: Devices & ASICs for Machine Intelligence and Post-Quantum Cryptography
Abstract: The high demand for AI services in conjunction with a dramatic chip shortage along with technology leaps such as 5/6G networks, cybersecurity threats, and quantum algorithms have resurrected a R&D push for advanced devices, information processing, and computing capability. To address this demand and explore novel technology, unique opportunities exist, for example, given by algorithmic parallelism of mixed-signal non-van Neuman accelerators. Especially electronic-photonic ASIC compute paradigms hold promise to enable non-iterative O(1) runtime complexity, ps-short latency, and TOPS/W throughputs. This opens prospects for next-generation hardware both for AI cloud services but also for accelerating edge computing such as enabled by compact and efficient PIC-CMOS co-designs pushing the SWAP envelope. As both a professor and a co-founder of a venture, in this seminar I will share my latest insights on fundamental complexity scaling and algorithm-hardware homomorphism on the one hand, and device- circuit- and system-level demonstrations on the other. I will introduce a novel memristive photonic RAM capable of zero-static power consumption suitable for AI edge applications and highlight our photonic tensor core ASIC demonstration leveraging parallelism including a software stack. Beyond matrix-matrix multiplication acceleration, I will show our Convolution Theorem-based accelerator enabling 1000x1000 matrix-size convolutions at 100us latency, or about 10x faster than today's GPUs. At the device level I will share advanced optoelectronics and quantum matter including a 50Gbps ITO-based modulator being 1,000x more compact than Silicon PDK solutions, discuss strainoptronic detectors with high gain-bandwidth-product, a 100GHz fast VCSEL, and share a path for an electrically-driven quantum source. Finally, having solved the complex-signal convolution I will show a Montgomery Multiplier for a data-center RSA public-key cryptosystem, and conclude by highlighting our recent post-quantum secure-hash-algorithm (SHA) system accelerating blockchain operations. I will conclude with an R&D outlook for the next decade and share examples of my passion supporting values and programs on diversity & inclusion.
Biography: Volker J. Sorger is an Associate Professor in the Department of Electrical and Computer Engineering and the Director of the Institute on AI & Photonics, the Head of the Devices & Intelligent Systems Laboratory at the George Washington University. His research areas include devices & optoelectronics, AI/ML accelerators, mixed-signal ASICs, quantum matter & processors, and cryptography. For his work, Dr. Sorger received multiple awards including the Presidential PECASE Award, the AFOSR YIP Award, the Emil Wolf Prize, and the National Academy of Sciences award of the year. Dr. Sorger is an Associate editor for OPTICA, serves on the board of Chip, and was the former editor-in-chief of Nanophotonics. He is a Fellow of Optica (former OSA), a Fellow of SPIE, a Fellow of the German National Academic Foundation, and a Senior Member of IEEE. He is a co-founder of Optelligence Company.
Host: ECE-Electrophysics
More Information: Volker Sorger Flyer.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Marilyn Poplawski
-
ECE-EP Seminar - Dejan Markovic, Thursday, March 24th at 10am in EEB 248 & via Zoom
Thu, Mar 24, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dejan Markovic, UCLA
Talk Title: The Future of Computing and Neuromodulation
Abstract: This talk will discuss future technologies addressing unmet needs in science, medicine, and engineering. Data-driven attentive computing requires runtime flexible and efficient hardware and software. Simple hardware leads to complex software (e.g. FPGA) and simple software leads to complex hardware (e.g. CPU). Runtime reconfigurable arrays (RTRAs) balance hardware and software to enable spatial and temporal flexibility for dynamic or uncertain environments. RTRA features multi-program tenancy, multi-size compile, and priority handling for >100x compute capacity gains over FPGA, and within 5x of (inflexible) hardware accelerators, as shown on a blind signal classification use case. Medical implants also require efficiency and flexibility, with heavily constrained size, weight and power, for novel clinical research and therapeutic systems. Despite notable clinical successes (e.g. Parkinson's disease), limitations in existing devices prevent them from expanding to other indications such as mental health or Alzheimer's disease. I will discuss the Neuro-stack, a versatile closed-loop system, verified in human subject experiments, towards miniaturized neural duplex of the future. These applications also reveal opportunities in system-level design automation to address design productivity and system assembly challenges.
Biography: Dejan MarkoviÄ is a Professor of Electrical and Computer Engineering at the University of California, Los Angeles (UCLA). He is also affiliated with UCLA Bioengineering Department, Neuroengineering field. He completed the Ph.D. degree in 2006 at the University of California, Berkeley, for which he was awarded 2007 David J. Sakrison Memorial Prize. His current research is focused on implantable neuromodulation systems, domain-specific compute architectures, and design methodologies. Dr. MarkoviÄ co-founded Flex Logix Technologies, a semiconductor IP startup, in 2014, and helped build foundational technology of Ceribell, a medical device startup. He received an NSF CAREER Award in 2009. In 2010, he was a co-recipient of ISSCC Jack Raper Award for Outstanding Technology Directions. He also received 2014 ISSCC Lewis Winner Award for Outstanding Paper. Prof. Markovic is a Fellow of the IEEE.
Host: ECE-Electrophysics
More Information: Dejan Markovic Flyer.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Marilyn Poplawski
-
ECE Seminar: Distributed Systems: Rigorous Theoretical Foundations Unlock Promising Gains
Fri, Mar 25, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Mohammad Ali Maddah-Ali, Research Scientist, Department of Electrical Engineering, Stanford University
Talk Title: Distributed Systems: Rigorous Theoretical Foundations Unlock Promising Gains
Abstract: Over the last twenty years, we have witnessed several revolutionary technologies, from communication networks to learning platforms to blockchains, that have profoundly changed our daily lives. Often, these platforms are modeled, designed, and operated based on intuition and folk wisdom. In this talk, we challenge some of those common beliefs. We show that by meticulously elaborating the key performance bottlenecks from first principles, we can propose counterintuitive solutions grounded in rigorous analysis that unlock considerable scaling gains in several areas:
1) In wireless communications, the delay in acquiring channel information is a significant bottleneck in supporting multiple users at a time. Contrary to popular belief, we demonstrate that even completely outdated channel information can be used for interference management and enabling simultaneous communications, thus alleviating the bottleneck of channel training.
2) In content delivery networks, folk wisdom design is to maximize the likelihood of serving a request from the local cache (hit rate); thus, the performance is bottlenecked by the size of an individual cache. We propose a fundamentally new approach with a gain that scales with the sum of the cache sizes in the network, rather than an individual cache size.
3) In distributed learning, we demonstrate that training with combined data samples (i.e., erasure-coded samples), rather than raw samples, can significantly improve the reliability and convergence rate. Moreover, we highlight the surprising role of approximation theory in circumventing a major bottleneck in designing practical coded training procedures.
We conclude with promising directions for further investigation: in particular, the challenges in adding decentralized trust and accountability to these systems, to place control over them back in the hands of individuals rather than big corporations.
Biography: Mohammad Ali Maddah-Ali received the B.Sc. degree from the Isfahan University of Technology, the M.Sc. degree from the University of Tehran, and the Ph.D. degree from the Department of Electrical and Computer Engineering, University of Waterloo, Canada. From 2008 to 2010, he was a Postdoctoral Fellow in the Department of Electrical Engineering and Computer Sciences, University of California at Berkeley. From 2010 to 2020, he was working at Bell Labs, Holmdel, NJ, as a communication network research scientist. He also worked as a faculty member at the Department of Electrical Engineering, Sharif University of Technology. Currently, he is a research scientist at the Department of Electrical Engineering, Stanford University.
Dr. Maddah-Ali is a recipient of several awards including the IEEE International Conference on Communications (ICC) Best Paper Award in 2014, the IEEE Communications Society and IEEE Information Theory Society Joint Paper Award in 2015, and the IEEE Information Theory Society Paper Award in 2016. He is currently serving as an associate editor for the IEEE Transactions on Information Theory and a lead editor for The IEEE Journal on Selected Areas in Information Theory.
Host: Dr. Keith Chugg, chugg@usc.edu
Webcast: https://usc.zoom.us/j/98149159985?pwd=cWFsVnRkZXRKcTlWYllMcy9Rempmdz09Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
WebCast Link: https://usc.zoom.us/j/98149159985?pwd=cWFsVnRkZXRKcTlWYllMcy9Rempmdz09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher