SUNMONTUEWEDTHUFRISAT
Conferences, Lectures, & Seminars
Events for February
-
Center of Autonomy and AI, Center for Cyber-Physical Systems and the Internet of Things, and Ming Hsieh Institute Seminar Series
Wed, Feb 02, 2022 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Jie Gao, Department of Computer Science, Rutgers University
Talk Title: Protecting Data Privacy in an Increasingly Connected World
Series: Center for Cyber-Physical Systems and Internet of Things
Abstract: Ubiquitous sensing, wireless communication and distributed computation have transformed the way we interact with the physical world. While we celebrate the convenience and the improved quality of life having smart devices around us, the data collection practices have moved from remote fields to private living environments and work spaces; and data collected have also shifted from non-sensitive scientific data to personal, sensitive data that is closely related to the user's health conditions, emotional states, physical activities and social relationships. In this talk I will review some of our recent work on protecting privacy of sensitive data by minimalist-style sensing and targeted data perturbation, as well as fundamental challenges/impossibility results regarding data privacy and utility.
Biography: Professor Jie Gao is currently Professor in Department of computer science, Rutgers University. She was on the faculty of Computer Science department at Stony Brook University from 2005-2019. She received B.Eng from the Special Class of the Gifted Young, University of Science and Technology of China in 1999, Ph.D in Computer Science from Stanford University in 2004 and was a postdoc at Caltech from 2004-2005. She received the NSF career award in 2006, IMC best paper award (2009), EWSN best paper award (2021) and multiple Research Excellence Awards in computer science department of Stony Brook. She is currently serving on the editorial board of ACM Transactions on Sensor Networks, International Journal of Computational Geometry and Applications and IEEE Transactions on Network Science and Engineering. She published over 140 referred papers in computer networking and theoretical computer science fields, and has graduated 17 Ph.D students.
Host: Pierluigi Nuzzo and Bhaskar Krishnamachari
Webcast: https://usc.zoom.us/webinar/register/WN_zyIBh_1gQLmKpMJG0GyLxwLocation: Online
WebCast Link: https://usc.zoom.us/webinar/register/WN_zyIBh_1gQLmKpMJG0GyLxw
Audiences: Everyone Is Invited
Contact: Talyia White
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Center of Autonomy and AI, Center for Cyber-Physical Systems and the Internet of Things, and Ming Hsieh Institute Seminar Series
Wed, Feb 09, 2022 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Shahriar Nirjon, Department of Computer Science at the University of North Carolina at Chapel Hill
Talk Title: Intermittent Learning on Harvested Energy
Series: Center for Cyber-Physical Systems and Internet of Things
Abstract: Years of technological advancements have made it possible for small, portable, electronic devices of today to last for years on battery power, and last forever - when powered by harvesting energy from their surrounding environment. Unfortunately, the prolonged life of these ultra-low-power systems poses a fundamentally new problem. While the devices last for years, programs that run on them become obsolete when the nature of sensory input or the operating conditions change. The effect of continued execution of such an obsolete program can be catastrophic. For example, if a cardiac pacemaker fails to recognize an impending cardiac arrest because the patient has aged or their physiology has changed, these devices will cause more harm than any good. Hence, being able to react, adapt, and evolve is necessary for these systems to guarantee their accuracy and response time. We aimed at devising algorithms, tools, systems, and applications that will enable ultra-low-power, sensor-enabled, computing devices capable of executing complex machine learning algorithms while being powered solely by harvesting energy. Unlike common practices where a fixed classifier runs on a device, we take a fundamentally different approach where a classifier is constructed in a manner that it can adapt and evolve as the sensory input to the system, or the application-specific requirements, such as the time, energy, and memory constraints of the system, change during the extended lifetime of the system.
Biography: Dr. Shahriar Nirjon is an Assistant Professor of Computer Science at the University of North Carolina at Chapel Hill, NC. He is interested in Embedded Intelligence -“ the general idea of which is to make resource constrained real-time and embedded sensing systems capable of learning, adapting, and evolving. Dr. Nirjon builds practical cyber-physical systems that involve embedded sensors and mobile devices, mobility and connectivity, and mobile data analytics. His work has applications in the area of remote health and wellness monitoring, and mobile health. Dr. Nirjon received his Ph.D. from the University of Virginia, Charlottesville, and has won a number of awards including four Best Paper Awards at Mobile Systems, Applications and Services (MOBISYS 2014), the Real-Time and Embedded Technology and Applications Symposium (RTAS 2012), Distributed Computing in Sensor Systems (DCOSS '19), and Challenges in AI and Machine Learning for IoT (AIChallengeIoT '20). Dr. Nirjon is a recipient of NSF CAREER Award in 2021. Prior to UNC, Dr. Nirjon has worked as a Research Scientist in the Networking and Mobility Lab at the Hewlett-Packard Labs in Palo Alto, CA.
Host: Pierluigi Nuzzo and Bhaskar Krishnamachari
Webcast: https://usc.zoom.us/webinar/register/WN_zyIBh_1gQLmKpMJG0GyLxwLocation: Online
WebCast Link: https://usc.zoom.us/webinar/register/WN_zyIBh_1gQLmKpMJG0GyLxw
Audiences: Everyone Is Invited
Contact: Talyia White
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE-S Seminar - Domain Specialized Architectures and Systems for AI/ML
Thu, Feb 10, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Divya Mahajan, Senior Researcher, Cloud Accelerated Systems & Technologies, Microsoft
Talk Title: Domain Specialized Architectures and Systems for AI/ML
Abstract: Advances in Artificial Intelligence (AI) and Machine Learning (ML) are beginning to revolutionize medicine, manufacturing, commerce, transportation, and other key aspects of our lives. However, such transformative effects are predicated on providing high-performance compute capabilities to enable these learning algorithms. Domain specific accelerators are an efficient and performant means to meet the compute requirements of these large-scale AI/ML. As the new age data-centers become heterogeneous with these emerging domain specific hardware, we must rethink both the architecture and the corresponding system stack.
In this talk, I will provide an overview of my contributions to design, deploy, and utilize accelerators for a wide class of AI/ML applications. I will first discuss pioneering works TABLA and DaNA, which are comprehensive full-stack solutions for machine learning accelerators that integrate with data management systems. These solutions expose a high-level programming interface to programmers that have limited knowledge about hardware design, nevertheless, can benefit from performance and efficiency gains through acceleration. Then, I will describe FAE, a novel framework that leverages statistical properties of data to best utilize the heterogeneous compute and memory resources for recommender model training. Finally, I will conclude with my future research vision towards devising architectures and systems for sustainable massive-scale distributed AI/ML by exploring the challenges which arise from the cross-pollination of different components in the data processing pipeline.
Biography: Divya Mahajan is a Senior Researcher in the Cloud Accelerated Systems & Technologies group at Microsoft. She leads the research, design, and deployment of communication primitives for massive-scale distributed deep learning. She obtained her PhD in Computer Science from Georgia Institute of Technology. She obtained her Masters from The University of Texas Austin, Texas and Bachelors from Indian Institute of Technology Ropar. Her research interests lie in designing novel architectures and building robust systems to address the needs of new and emerging applications. She is passionate about continuing innovative research to have a broad impact on computing and society in general.
Divya is the recipient of National Council for Women and Information Technology Collegiate Award, President of India Gold Medal at IIT, and has been a Finalist in the Qualcomm Innovation Fellowships. Her work has been recognized with the College of Computing Dissertation Award, HPCA Distinguished Paper Award, and has appeared in top architecture, database, systems, and machine learning venues like ISCA, MICRO, HPCA, ASPLOS, VLDB, NeurIPS and high impact journals like IEEE Micro.
https://www.microsoft.com/en-us/research/people/divyam/
Host: Dr. Murali Annavaram, annavara@usc.edu
Webcast: https://usc.zoom.us/j/96503892197?pwd=Nk13S1RZb25tMlN1QnUxRWZXN2lNZz09More Information: ECE Seminar-Mahajan-021022.pdf
WebCast Link: https://usc.zoom.us/j/96503892197?pwd=Nk13S1RZb25tMlN1QnUxRWZXN2lNZz09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE Seminar - Dealing with Data Deluge in the Edge Systems
Mon, Feb 14, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Ramyad Hadidi, Machine Learning Researcher, SK hynix America
Talk Title: Dealing with Data Deluge in the Edge Systems
Abstract: Each day, a huge amount of data is generated. Edge systems--any computing agent but large-scale datacenter machines--not only fuel this data deluge but also play an increasingly vital role in processing them. Unlike conventional systems that are engineered with abundant resources, edge systems operate in real-world conditions facing several unknown design-space trade-offs with limited resources restricting their full-scale autonomy. This leads to isolated, time-consuming, and costly approaches to each challenge that result in ad-hoc edge systems but not the optimal one. To effectively maneuver the constraints and unique multi-dimensional design space of edge systems, my research develops novel machine learning techniques and exploits hardware-software synergy by setting roadmaps across the hardware-software stack for the next generation of edge systems.
In my talk, first, with an example of quadcopter drones, I show how my research is a pioneer that reveals the unique multi-dimensional design space of edge systems and suggests optimal points within this space depending on the use case. By formulating the fundamental drone subsystems and introducing our open-source customizable drone, I explain how computations impact this design space. As an example of optimized computations, by exploring implementations of simultaneous localization and mapping (SLAM) on various hardware platforms (CPU, GPU, FPGA, and ASIC), I demonstrate which implementation is more reasonable for drones. The second part of my talk emphasizes the necessity of modern machine learning techniques, such as those utilizing heavy neural networks, in comprehending complex raw data in edge systems and acting upon the outcomes. I show how my research empowers edge devices to break their individual resource constraints by distributing the computation on collaborating peer devices and proposes edge-aware neural networks by exploring hardware-software co-designs, algorithmic modifications, and system-level optimizations. In the end, I propose my plans for effectively handling data in exotic frontiers of edge systems with unique constraints to stimulate thought-provoking applications for our future.
Biography: Ramyad Hadidi is currently a machine learning researcher at SK hynix working at the intersection of hardware, software, and edge devices, focusing on the efficient execution of deep learning algorithms. Ramyad Hadidi received his Ph.D. in computer science from Georgia Institute of Technology in May 2021 under the supervision of Professor Hyesoon Kim with his thesis titled "Deploying Deep Neural Networks in Edge with Distribution." Ramyad's research interests include but are not limited to computer architecture, robotics, edge computing, and machine learning. Besides his dissertation research, Ramyad has contributed to research on processing-in-memory, GPU systems, and hardware accelerators for sparse problems, believing a balance between depth and breadth leads to genuine research problems.
Host: Dr. Peter Beerel, pabeerel@usc.edu
Webcast: https://usc.zoom.us/j/93601839133?pwd=QW51RjRSVSsyZStDOVk2RUZ3Q0ZUdz09WebCast Link: https://usc.zoom.us/j/93601839133?pwd=QW51RjRSVSsyZStDOVk2RUZ3Q0ZUdz09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE Seminar - Scalable and Trustworthy Learning for Distributed Intelligence
Wed, Feb 16, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Tianyi Chen, Assistant Professor, Department of Electrical, Computer, and Systems Engineering, Rensselaer Polytechnic Institute
Talk Title: Scalable and Trustworthy Learning for Distributed Intelligence
Abstract: The past decade has witnessed the revival of artiï¬cial intelligence (AI) and machine learning (ML) in almost every branch of science and technology. The "fuel" to AI and ML is supplied by the surge of data and computing power. Today, data and computing power are distributed among wireless devices and companies that we term data owners. Due to the pressing need for data in AI/ML tasks and the increasing concerns on data privacy, a sizeable number of AI/ML tasks will be executed across networked data owners with the vision of distributed intelligence.
In this talk, I will use federated learning (FL) as an example of distributed intelligence. I will highlight its key challenges when it interacts with wireless networks such as efficiency, privacy, security, and robustness. I will focus on two aspects of scalable and trustworthy FL - efficiency and privacy. From a unified view of information correlation among iterative FL updates, I will elaborate i) how we can leverage such correlation to improve the efficiency of FL; and ii) how such correlation may be leveraged by a malicious third party to risk data privacy. Our methods are simple to implement, and they come with rigorous performance guarantees. I will conclude this talk by highlighting a few directions that I will pursue towards distributed intelligence beyond FL.
Biography: Tianyi Chen is an Assistant Professor in the Department of Electrical, Computer, and Systems Engineering at Rensselaer Polytechnic Institute (RPI), where he is jointly supported by the RPI - IBM Artificial Intelligence Research Partnership. Dr. Chen received his B. Eng. degree in Electrical Engineering from Fudan University in 2014, and the Ph.D. degree in Electrical and Computer Engineering from the University of Minnesota in 2019. During 2017-2018, he has been a visiting scholar at Harvard University, University of California, Los Angeles, and University of Illinois Urbana-Champaign. Dr. Chen's research focuses on theoretical and algorithmic foundations of optimization, machine learning, and statistical signal processing, with applications in networked computing systems such as wireless and IoT systems.
Dr. Chen is the inaugural recipient of IEEE Signal Processing Society Best PhD Dissertation Award in 2020 and a recipient of NSF CAREER Award in 2021. He is also the co-author of several best paper awards such as the Best Student Paper Award at the NeurIPS Federated Learning Workshop in 2020 and at IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) in 2021.
Host: Dr. Salman Avestimehr, avestime@usc.edu
Webcast: https://usc.zoom.us/j/98177654001?pwd=R1h6bUJZUXcxZENZYWtVYmorRVNFQT09WebCast Link: https://usc.zoom.us/j/98177654001?pwd=R1h6bUJZUXcxZENZYWtVYmorRVNFQT09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Center of Autonomy and AI, Center for Cyber-Physical Systems and the Internet of Things, and Ming Hsieh Institute Seminar Series
Wed, Feb 16, 2022 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Changliu Liu, Robotics Institute, Carnegie Mellon University
Talk Title: Safe Control and Learning for Effective Human-Robot Collaboration
Series: Center for Cyber-Physical Systems and Internet of Things
Abstract: In this talk, I will discuss our recent work on safe control and learning for effective human-robot collaboration. I will first introduce safe control methods using energy-function-based methods, then discuss how to combine them with learning controllers where an explicit analytical dynamic model of the system is usually not available (especially in human-robot systems). These safe control methods will enable safe reinforcement learning with zero training time violation. Then I will discuss about methods to robustly learn models to predict human behaviors. The key challenge we need to address is the distribution shift between the offline collected human behavioral data and the online measured human behaviors. To mitigate the distribution shift, we introduce two methods: online model adaptation, and offline verification-guided data augmentation. These methods have been applied to facilitate human-robot collaboration in industrial assembly tasks. I will conclude the talk with future visions on how to effectively deploy human-robot systems on factory floors.
Biography: Dr. Changliu Liu is an assistant professor in the Robotics Institute, School of Computer Science, Carnegie Mellon University (CMU), where she leads the Intelligent Control Lab. Prior to joining CMU, Dr. Liu was a postdoc at Stanford Intelligent Systems Laboratory. She received her Ph.D. from University of California at Berkeley and her bachelor's degrees in engineering and economics from Tsinghua University. Her research interests lie in the design and verification of intelligent systems with applications to manufacturing and transportation. She published the book "Designing robot behavior in human-robot interactions" with CRC Press in 2019. She received many best paper awards, Rising Star in EECS, NSF Career Award, Amazon Research Award, and Ford URP Award.
Host: Pierluigi Nuzzo and Somil Bansal
Webcast: https://usc.zoom.us/webinar/register/WN_zyIBh_1gQLmKpMJG0GyLxwLocation: Online
WebCast Link: https://usc.zoom.us/webinar/register/WN_zyIBh_1gQLmKpMJG0GyLxw
Audiences: Everyone Is Invited
Contact: Talyia White
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE Seminar - Human-centered machine intelligence: From robust signal analytics to trustworthy human-technology partnership
Thu, Feb 17, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Theodora Chaspari, Assistant Professor, Computer Science & Engineering Department, Texas A&M University
Talk Title: Human-centered machine intelligence: From robust signal analytics to trustworthy human-technology partnership
Abstract: Recent converging advances in sensing and computing allow the ambulatory long-term tracking of individuals yielding a rich set of real-life multimodal bio-behavioral signals, such as speech, physiology, and facial expressions. While such measurements coupled with artificial intelligence (AI) and machine learning (ML) algorithms have been heralded as promising solutions to addressing pressing societal challenges, public and expert determination of whether this integration is a good prospect is widely debated. At the same time, interactions between humans and AI are increasingly moving away from simple diagnosis of human outcomes to collaborative relationships, in which humans work side-by-side with AI systems for carrying out a set of common goals. This talk will describe new signal analytics and ML algorithms for trustworthy human-centered machine intelligence focusing on four main pillars of trustworthiness, namely robustness, privacy preservation, explainability, and fairness. We will first present our work on personalized ML models of human outcomes, generalizable learning of human states via the formulation of weakly supervised learning algorithms, and context-aware signal representations for reliably modeling interpersonal interaction. Following that, we will discuss a privacy-preserving mood recognition framework through user anonymization and examine factors of socio-demographic bias in signals and ML systems that may perpetuate social disparities in human-centered analytics. Finally, we will present our recent work on human-AI collaboration that examines how human stakeholders (e.g., clinicians) interact with AI/ML along dimensions of trust formation, maintenance, and repair. We will demonstrate the effectiveness of the proposed approaches through examples in mental health, public health, workforce training and re-skilling, and team science.
Biography: Theodora Chaspari is an Assistant Professor in the Computer Science & Engineering Department at Texas A&M University. She has received her Bachelor of Science (2010) in Electrical & Computer Engineering from the National Technical University of Athens, Greece and her Master of Science (2012) and Ph.D. (2017) in Electrical Engineering from the University of Southern California. Between 2010 and 2017 she worked as a Research Assistant at the Signal Analysis and Interpretation Laboratory at USC. She has also been a Lab Associate Intern at Disney Research (2015). Theodora's research interests lie in the areas of signal processing, machine learning, data science, and affective computing. She is a recipient of the NSF CAREER Award (2021), TAMU Montague Teaching Award (2021), USC Women in Science and Engineering Merit Fellowship (2015), and USC Annenberg Graduate Fellowship (2010). Papers co-authored with her students have been nominated and won awards at the ASC 2021, ACM BuildSys 2019, IEEE ACII 2019, ASCE i3CE 2019, and IEEE BSN 2018 conferences. She is serving as an Editor of the Elsevier Computer Speech & Language, and in various conference organization committees (ACM ICMI 2023/2020/2018, ACM IUI 2021, ACM KDD 2022, IEEE ACII 2022/2021/2019/2017, IEEE BSN 2018). She has further developed and taught several graduate and undergraduate courses in signal analytics and ML. Her work is supported by federal and private funding sources, including the NSF, NIH, NASA, IARPA, AFRL, AFOSR, General Motors, Keck Foundation, and the Engineering Information Foundation.
Host: Dr. Justin Haldar, jhaldar@usc.edu
Webcast: https://usc.zoom.us/j/91654081503?pwd=aTFTN293KzFMTWFIUlQ4MkJHOFQxdz09Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
WebCast Link: https://usc.zoom.us/j/91654081503?pwd=aTFTN293KzFMTWFIUlQ4MkJHOFQxdz09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE Seminar
 Lin.jpg)
Fri, Feb 18, 2022 @ 10:00 AM - 11:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Yingyan (Celine) Lin, Department of Electrical and Computer Engineering, Rice University
Talk Title: Towards Network-Accelerator Co-Search for Promoting Ubiquitous on-Device Intelligence and Green AI
Abstract: Deep learning (DL)-powered intelligence embedded into numerous daily-life devices promises to transform the quality of human life. Despite this great promise, there is a vast and increasing gap between the prohibitive complexity of powerful DL algorithms and the constrained resources in daily-life devices. While DL accelerators have the potential to close the aforementioned immense gap and push forward green AI, their power has yet to be unleashed due to the following fundamental challenges: (1) fast DL algorithm advances vs. slow DL accelerator development, and (2) the promise of algorithm and accelerator co-search vs. the Lack of such co-search. Therefore, it is imperative to develop innovative techniques that can expedite the development of optimal DL accelerators and unlock the promise of co-searching for optimal DL algorithms and accelerators for maximizing their achievable hardware efficiency.
In this talk, I will present our recently developed techniques towards DL network-accelerator co-search, serving as a timely holistic effort toward addressing the aforementioned challenges. Specifically, I will start by introducing our techniques for designing hardware-aware DL algorithms (i.e., top-down efforts) and algorithm-award DL accelerators (i.e., bottom-up efforts), which helps us to gain important insights for understanding their design space and optimization. Then, I will share our first-of-their-kind techniques that are among the very first generic efforts to enable simultaneous searching for optimal DL algorithms and accelerators (i.e., bridging efforts) to maximize both task accuracy and hardware efficiency. Finally, I will conclude my talk with exciting (1) applications of our co-search framework and (2) pointers to future directions.
Biography: Yingyan (Celine) Lin is an Assistant Professor in the Department of Electrical and Computer Engineering at Rice University. She leads the Efficient and Intelligent Computing (EIC) Lab at Rice, which focuses on developing efficient machine learning techniques towards green AI and ubiquitous machine learning powered intelligence. She received a Ph.D. degree in Electrical and Computer Engineering from the University of Illinois at Urbana-Champaign in 2017.
Prof. Lin is a NSF CAREER Award, IBM Faculty Award, and Facebook Research Award recipient. She was selected as a Rising Star in EECS by the 2017 Academic Career Workshop for Women at Stanford University. She received a Best Student Paper Award at the 2016 IEEE International Workshop on Signal Processing Systems (SiPS 2016), and the 2016 Robert T. Chien Memorial Award for Excellence in Research at UIUC. Prof. Lin is currently the lead PI on multiple multi-university projects (e.g., RTML and 3DML) and her group has been funded by NSF, NIH, DARPA, ONR, Qualcomm, Intel, IBM, and Facebook.
Host: Dr. Murali Annavaram, annavara@usc.edu
Webcast: https://usc.zoom.us/j/95976355759?pwd=cTkwMlpoVFRuOEkxQ2JHNy9MYzJ4dz09WebCast Link: https://usc.zoom.us/j/95976355759?pwd=cTkwMlpoVFRuOEkxQ2JHNy9MYzJ4dz09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE Seminar: Next-Generation Wireless Networks for Billions of IoT Devices
Tue, Feb 22, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Ali Abedi, Research Lecturer, Cheriton School of Computer Science, University of Waterloo
Talk Title: Next-Generation Wireless Networks for Billions of IoT Devices
Abstract: It is estimated that the total number of Internet of Things (IoT) devices will grow to 80 billion devices in a few years. Unfortunately, existing wireless networks cannot satisfy the diverse requirements of IoT applications in terms of power consumption, data rate, and privacy. Some IoT devices, such as contact sensors used in intrusion detection systems, transmit only a few bytes of data occasionally, while other devices such as virtual reality headsets require a continuous stream of data with a high data rate. Moreover, many IoT devices run on a battery; therefore, they have very strict power consumption requirements. The battery in many IoT devices has to be changed every few months, which is time consuming and costly. Furthermore, old batteries have adverse environmental effects, if not disposed of properly. In addition to these problems, bringing many IoT devices to our smart homes and offices creates many privacy concerns. How can users be confident that their privacy is not violated in smart environments?
In this talk, I present the design of next generation wireless networks that satisfy the diverse requirements of IoT applications. To enable low-power wireless networking for IoT applications that require low data rates, I present a system that enables a battery-free IoT device to transmit its data to nearby WiFi devices. Next, I describe the design of a low-power and low-cost millimeter wave network for IoT devices that require up to 100 Mbps of bandwidth. Finally, I discuss privacy issues caused by wireless signals transmitted by many IoT devices in a smart environment.
Biography: Ali Abedi is currently a research lecturer at the University of Waterloo. His research interests are in the areas of wireless networks and mobile systems with a special focus on the Internet of Things (IoT) and smart environments. He received his Ph.D. in computer science from the University of Waterloo. His work has been published in top systems and networking venues such as SIGCOMM, MobiCom, and HotNets. He was awarded the gold medal in the Student Research Competition (SRC) competition at Mobicom 2018. His research projects have been featured in ACM GetMobile, ACM Tech News, and Science Daily. He has received multiple grants from the Natural Sciences and Engineering Research Council of Canada (NSERC). His research has resulted in multiple patents, and has attracted interests from companies such as Google, Qualcomm, and ecobee.
Host: Dr. Konstantinos Psounis, kpsounis@usc.edu
Webcast: https://usc.zoom.us/j/96468306783?pwd=cmJGWE91d1M0VDM1aGhaaXJNdDFPZz09WebCast Link: https://usc.zoom.us/j/96468306783?pwd=cmJGWE91d1M0VDM1aGhaaXJNdDFPZz09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE Seminar: Solving the Cloud Efficiency Crisis with Fast and Accessible Scheduling
Tue, Feb 22, 2022 @ 12:00 PM - 01:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Kostis Kaffes, Electrical Engineering Department, Stanford University
Talk Title: Solving the Cloud Efficiency Crisis with Fast and Accessible Scheduling
Abstract: Operating systems (OS) specialization is necessary as the one-size-fits-all approach of fundamental OS operations such as scheduling is incompatible with today's diverse application landscape. Such specialization can improve application performance and cloud platform efficiency by an order of magnitude or more. Towards this goal, I will first discuss Shinjuku, a specialized OS that supports an order of magnitude higher load and lower tail latency than state-of-the-art systems by enabling better scheduling. Shinjuku leverages hardware support for virtualization to preempt as often as every 5 microseconds and disproves the conventional wisdom that interrupts are incompatible with microsecond timescales. Then, I will present Syrup, a framework that enables everyday application developers to specify custom scheduling policies easily and safely deploy them across different layers of the stack over existing operating systems like Linux, bringing the benefits of specialized scheduling to everyone. For example, Syrup allowed us to implement policies that previously required specialized dataplanes in less than 20 lines of code and improve the performance of an in-memory database by 8x without needing any application modification.
Biography: Kostis Kaffes is a final-year Ph.D. candidate in Electrical Engineering at Stanford University, advised by Christos Kozyrakis. He is broadly interested in computer systems, cloud computing, and scheduling. His thesis focuses on end-host, rack-scale, and cluster-scale scheduling for microsecond-scale tail latency with the goal of improving efficiency in the cloud. Recently, he has been looking for ways to make it easier to implement and deploy custom scheduling policies across different layers of the stack. Kostis's research has been supported by a Facebook Research Award and various scholarships and fellowships from Stanford, A.G. Leventis Foundation, and Gerondelis Foundation. Prior to Stanford, he received his undergraduate degree in Electrical and Computer Engineering from the National Technical University of Athens in Greece.
Host: Dr. Murali Annavaram, annavara@usc.edu
Webcast: https://usc.zoom.us/j/96988520485?pwd=aHRIY1BBWW5PVEtCeDlWSnAwUUxsUT09Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
WebCast Link: https://usc.zoom.us/j/96988520485?pwd=aHRIY1BBWW5PVEtCeDlWSnAwUUxsUT09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE Seminar: Human/System Co-design to Protect Data Privacy
Wed, Feb 23, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Haojian Jin, PhD Candidate, Human-Computer Interaction Institute, Carnegie Mellon University
Talk Title: Human/System Co-design to Protect Data Privacy
Abstract: Privacy is changing how we build computing systems. Recent regulations, such as General Data Protection Regulation, California Consumer Privacy Act, Children's Online Privacy Protection Act, require developers to offer greater privacy protections. However, developers struggle to turn these high-level privacy principles into low-level code implementation.
The primary cause of this difficulty is that privacy is a multi-stakeholder issue: developers want to achieve more functionality and productivity; users want more control with lower effort; regulators wish to audit systems with limited resources and do not want to stifle innovation; finally, system deployments need to remain proprietary and efficient.
In this talk, I will present two systems to illustrate that these Human/System requirements can jointly inform system design up-front and not be afterthoughts. I will describe (1) applying human/system co-design for data minimization, a foundational privacy principle in modern privacy regulation, and (2) how user and other stakeholder experience is transformed in co-designed systems. I will conclude with plans to create a virtuous cycle ecosystem where building trustworthy systems is rewarded, and developers compete to guarantee greater user protection, not less.
Biography: Haojian Jin is a final-year Ph.D. candidate in the Human-Computer Interaction Institute at Carnegie Mellon University, advised by Jason Hong and Swarun Kumar. His research lies at the intersection of human-computer interaction, privacy, and mobile systems. His work has been recognized with a UbiComp Gaetano Borriello Outstanding Student Award, Research Highlights at Communications of the ACM and GetMobile, and best paper awards at Ubicomp and ACM Computing Reviews. See more at: http://haojianj.in/.
Host: Dr. Bhaskar Krishnamachari, bkrishna@usc.edu
Webcast: https://usc.zoom.us/j/92527250101?pwd=dlQ1YzV1enJTYnRaQmFBbFpnZS9ZQT09Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
WebCast Link: https://usc.zoom.us/j/92527250101?pwd=dlQ1YzV1enJTYnRaQmFBbFpnZS9ZQT09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Center of Autonomy and AI, Center for Cyber-Physical Systems and the Internet of Things, and Ming Hsieh Institute Seminar Series
Wed, Feb 23, 2022 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Gaurav Gupta, Amazon Web Services (AWS) AI lab
Talk Title: Operator Learning for Partial Differential Equations
Series: Center for Cyber-Physical Systems and Internet of Things
Abstract: The partial differential equations (PDEs) model several real-world setups of Physics, Engineering, biology, Epidemiology. The solution can be formulated as an operator map problem. We show that learning the operator kernels can be efficiently performed by exploiting the fundamental properties. We will discuss a novel multiwavelets-based neural operator approach to achieve a compressed representation and show applications on several benchmarks PDE datasets. Next, we also discuss a class of PDEs called 'Initial Value Problems,' which has applications in predictions and forecasting. We develop a compact non-linear neural operator which maps initial conditions to activities at a later time. The proposed approach yields data efficiency which is necessary to deal with scarce real-world datasets, and as a case study we formulate and solve urgent real-world problems like Epidemic forecasting (e.g., COVID19).
Biography: Gaurav Gupta is currently a researcher (Applied Scientist) at Amazon Web Services (AWS) AI labs. He completed his PhD from USC Viterbi. His research interests span the domain of time-series modeling, learning partial differential equations, information theory for machine learning, fractional dynamical models, complex networks, brain EEG signals modeling. He is working on inter-disciplinary mathematical and applied problems on forecasting, PDEs, and has publications in top venues like Neurips, ICLR, Nature, IEEE Control Society, ACM cyber-physical society.
Host: Pierluigi Nuzzo, nuzzo@usc.edu
Webcast: https://usc.zoom.us/webinar/register/WN_zyIBh_1gQLmKpMJG0GyLxwLocation: Online
WebCast Link: https://usc.zoom.us/webinar/register/WN_zyIBh_1gQLmKpMJG0GyLxw
Audiences: Everyone Is Invited
Contact: Talyia White
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE-S Seminar: Faster, smarter, and greener systems for data-center scale AI
Thu, Feb 24, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Udit Gupta, PhD Candidate, Harvard University
Talk Title: Faster, smarter, and greener systems for data-center scale AI
Abstract: The modern Internet is driven by AI-centric services that determine how we interact with technology and society on a daily basis. The exponential rise in AI is largely fueled by the design, development, and deployment of domain-specific software and hardware that have yielded orders of magnitude improvements for deep learning. Despite these efforts, this talk focuses on an important, yet under-studied area: systems for deep learning-based personalized recommendation. Personalized recommendations form the backbone of our interaction with the Internet including search, e-commerce, streaming, and social media. Systems play a crucial role in enabling accurate, efficient, and sustainable recommendation engines.
In this talk I show how modern deep learning-based personalized recommendation engines not only consume the majority of AI training and inference cycles in production data centers, but also introduce unique system design challenges to efficient execution. To tackle these challenges, I design solutions across the software and hardware stack to optimize inference efficiency by jointly considering application-level characteristics, unique neural network model architectures, data-center scale implications, and the underlying hardware. Given the rapidly growing infrastructure demands posed by AI and recommendation engines, my work highlights that systems must go beyond performance, power, and energy efficiency to consider environmental footprint as a first order design target to enable sustainable computing. Finally, I chart paths to designing future systems that enable emerging AI-driven applications by balancing performance, efficiency, sustainability, and privacy.
Biography: Udit Gupta is a PhD student at Harvard University and visiting research scientist at Facebook AI Research. His research interests focus on enabling next-generation responsible AI platforms by designing novel computer systems and hardware. His recent work focuses on the optimization of data center-scale deep learning-based personalized recommendation engines (HPCA 2020, ISCA 2020, MICRO 2021, ASPLOS 2021) and enabling sustainable computing by considering the environmental impact of end-to-end hardware life cycles (HPCA 2021, MLSys 2022). Udit's work has been evaluated at-scale in production data centers and incorporated into standardized benchmarks and infrastructure used by the research community. His research has been recognized as an IEEE MICRO Top Picks honorable mention in 2020 and received an IEEE MICRO Top Picks award in 2021, as well as nominated for best paper at PACT 2019 and DAC 2018. In addition to research, Udit is passionate about building interdisciplinary communities. He has co-founded the PeRSonAl (personalized recommendation systems and algorithms) workshop and CLEAR (computing landscapes with environmental accountability and responsibility) workshops co-located at systems and machine learning conferences like ASPLOS, ISCA, and MLSys. He is also the co-chair of the Computer Architecture Student Association.
Host: Dr. Murali Annavaram, annavara@usc.edu
Webcast: https://usc.zoom.us/j/96028058998?pwd=cFFFSm1rdjFBcjdiMURMOWpxMi9tUT09Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
WebCast Link: https://usc.zoom.us/j/96028058998?pwd=cFFFSm1rdjFBcjdiMURMOWpxMi9tUT09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
CILQ Internal Seminar
Fri, Feb 25, 2022 @ 12:00 PM - 01:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Alan Willner, Professor, USC
Talk Title: Optical Communications: Innovations and Applications Abound
Abstract: Optical communications has enjoyed tremendous impact over the past 50 years. Relatively soon after the concrete proposal of optical fiber communications was reported and the low-loss fiber was demonstrated, fiber-based communications dramatically impacted the way society transfers information. However, there are other key areas beyond fiber-based communications that were also envisioned ~50 years ago but are only recently emerging. Such emergence is due to enhanced capacity needs and critical innovations, including advances in photonic integrated circuits (PICs). This talk will highlight various examples of the innovations and emerging applications of optical communications, including:
1. Free-space optical communications: As opposed to RF, optical links have high directionality and large bandwidth. There is great excitement in the recent emergence of deployed free-space optical links, be they through air or outer-space. Moreover, due to the extremely high losses of RF, even underwater links in the blue-green are gaining significant interest. Also to be discussed is capacity enhancement using multiplexing of multiple orbital-angular-momentum beams.
2. Non-conventional wavelengths: Fiber systems are overwhelmingly in the near-IR, whereas free-space links can take advantage of a much wider frequency range, from THz to visible. Such systems may utilize: (a) native high-speed components, and/or (b) wavelength-band conversion of near-IR channels to other frequencies.
3. Optical signal processing (OSP): OSP has long held the promise of high-speed operation and the avoidance of inefficient optical-electrical-optical conversion. Although OSP deployment has been limited, advances in PICs, power efficiency and multi-wavelength operation may soon enable the emergence of OSP for high-performance functions.
Host: CILQ
Webcast: https://usc.zoom.us/j/92417517950?pwd=WUkycy90cndVQko5R3RhQ1U3STBDdz09Location: via zoom
WebCast Link: https://usc.zoom.us/j/92417517950?pwd=WUkycy90cndVQko5R3RhQ1U3STBDdz09
Audiences: Everyone Is Invited
Contact: Corine Wong
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ECE Seminar: Certifiable Outlier-Robust Geometric Perception: Robots that See through the Clutter with Confidence
Mon, Feb 28, 2022 @ 10:00 AM - 11:00 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Heng Yang, Laboratory for Information & Decision Systems, Department of Mechanical Engineering, MIT
Talk Title: Certifiable Outlier-Robust Geometric Perception: Robots that See through the Clutter with Confidence
Abstract: Geometric perception is the task of estimating geometric models (e.g., object pose and 3D structure) from sensor measurements and priors (e.g., point clouds and neural network detections). Geometric perception is a fundamental building block for robotics applications ranging from intelligent transportation to space autonomy. The ubiquitous existence of outliers -measurements that tell no or little information about the models to be estimated- makes it theoretically intractable to perform estimation with guaranteed optimality. Despite this theoretical intractability, safety-critical robotics applications still demand trustworthiness and performance guarantees on perception algorithms. In this talk, I present certifiable outlier-robust geometric perception, a new paradigm to design tractable algorithms that enjoy rigorous performance guarantees, i.e., they return an optimal estimate with a certificate of optimality for a majority of problem instances, but declare failure and provide a measure of suboptimality for worst-case instances. Particularly, I present two general-purpose algorithms in the certifiable perception toolbox: (i) an estimator that uses graph theory to prune gross outliers and leverages graduated non-convexity to compute the optimal model estimate with high probability of success, and (ii) a certifier that employs sparse semidefinite programming (SDP) relaxation and a novel SDP solver to endow the estimator with an optimality certificate or escape local minima otherwise. The estimator is fast and robust against up to 99% random outliers in practical perception applications, and the certifier can compute high-accuracy optimality certificates for large-scale problems beyond the reach of existing SDP solvers. I showcase certifiable outlier-robust perception on robotics applications such as scan matching, satellite pose estimation, and vehicle pose and shape estimation. I conclude by remarking three opportunities arising from certifiable perception: to speedup online global optimization by offline learning from data; to enable safe learning-based perception by bridging certifiable estimation with deep representation learning; and to couple and unify perception with action towards trustworthy autonomy.
Biography: Heng Yang is a final-year Ph.D. candidate in the Laboratory for Information & Decision Systems and the Department of Mechanical Engineering at the Massachusetts Institute of Technology (MIT), working with Prof. Luca Carlone. He holds a B.S. degree from Tsinghua University and an S.M. degree from MIT, both in Mechanical Engineering. His research interests include large-scale convex optimization, semidefinite relaxation, robust estimation, and machine learning, applied to robotics and trustworthy autonomy. His work includes developing certifiable outlier-robust machine perception algorithms, large-scale semidefinite programming solvers, and self-supervised geometric perception frameworks. Heng Yang is a recipient of the Best Paper Award in Robot Vision at the 2020 IEEE International Conference on Robotics and Automation (ICRA), a Best Paper Award Honorable Mention from the 2020 IEEE Robotics and Automation Letters (RA-L), and a Best Paper Award Finalist at the 2021 Robotics: Science and Systems (RSS) conference. He is a Class of 2021 RSS Pioneer.
Host: Dr. Keith Chugg, chugg@usc.edu
Webcast: https://usc.zoom.us/j/91553052387?pwd=V0NqTFNJMlBNZkxWVnVIQmYrVWtVQT09Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
WebCast Link: https://usc.zoom.us/j/91553052387?pwd=V0NqTFNJMlBNZkxWVnVIQmYrVWtVQT09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor.
 Lin.jpg)
