November 5th – Memory and Communication Walls in the Exascale Era of Computing

This week, we have a presentation by Mert Hidayetoglu, a PhD candidate in the area of high performance computing.

Title:
Memory and Communication Walls in the Exascale Era of Computing

Abstract:
As we are going into the exascale era of computing, we are racing to demonstrate technological capabilities for solving critical problems for the well-being of citizens of all countries today, such as understanding climate change or simulating COVID-19 mechanisms at molecular level. In fact, the exascale computer merely celebrates a psychological milestone because only a handful of applications will enjoy exascale deployment on early systems. Because, a plethora of today’s relevant scientific, AI, and graph-analytics workloads involve irregular and sparse computational patterns, and they therefore suffer from memory-wall and communication-wall bottlenecks. As a result, these applications utilize only a tiny portion of the theoretical performance of large-scale computing systems in practice. In this talk, I will explain memory-wall and communication-wall bottlenecks, and present Tiled SpMM and hierarchical communication techniques for overcoming those on supercomputers with multi-GPU node architecture.

Bio:
Mert Hidayetoglu is a PhD candidate at ECE Illinois with research at the intersection of large-scale applications, high-performance computing, and software systems. His dissertation focuses on addressing inefficiencies with unstructured data accesses, communications, and computations on supercomputers with multi-GPU node architecture. His SC20 paper on X-ray imaging won the best paper award and his HPEC’20 paper on sparse DNN inference won MIT/Amazon/IEEE Graph Challenge. He is a 2021 recipient of ACM/IEEE-CS George Michael Memorial HPC Fellowship.

October 29th – Health Care Engineering Systems Center

This week, we have a presentation by Professor Thenkurussi “Kesh” Kesavadas – Director of the Health Care Engineering Systems Center at University of Illinois at Urbana-Champaign.

Title:
Health Care Engineering Systems Center (HCESC) at the University of Illinois at Urbana-Champaign

Abstract:
The Health Care Engineering Systems Center (HCESC) at the University of Illinois at Urbana-Champaign provides clinical immersion and fosters collaboration between engineers and physicians. Research areas include simulation technologies, smart health systems, data analytics, human factors, and medical robotics. One of the most important activities of HCESC is managing the Jump Applied Research in Community Health through Engineering and Simulation (ARCHES) endowment, which was established in 2014 by a $62.5 million gift. Since its inception, Jump ARCHES has provided over $ 5 million through an internal grant mechanism to fund over 93 proposals. In this presentation we will discuss center’s research and educational mission and provide information about how CSL faculty and students can participate in the center activities.

Bio:
Thenkurussi “Kesh” Kesavadas is the founder Director of the Health Care Engineering Systems Center at the University of Illinois at Urbana-Champaign. He is a professor of Industrial and Enterprise Systems Engineering, and holds faculty appointments in the Department of Computer Science, the Department of Electrical and Computer Engineering and with the Carle-Illinois College of Medicine at the university. His current research focuses on medical robotics and virtual reality for medicine. Kesavadas has widely published in the areas of medical robotics, VR, medical simulation and manufacturing. He is also an inventor of products that were successfully commercialized. In 2008, Kesavadas developed the world’s first stand-alone virtual reality Robotic Surgical Simulator called RoSSTM that is used around the world to train residents and medical students. This invention led to forming Simulated Surgical Systems to commercialize RoSSTM. Before joining Illinois, Kesavadas was a professor in the Department of Mechanical and Aerospace Engineering at the State University of New York at Buffalo (NY) where, in 1996, he founded one of the first Virtual Reality labs in the US. Kesavadas received his Ph.D. from the Pennsylvania State University, an M.S. from Indian Institute of Technology, Madras, and a bachelor’s degree from the University of Calicut. stitute of Technology, Madras, and a bachelor’s degree from the University of Calicut.

October 22nd – RL Based Disease Progression Model for Alzheimer’s Disease

This week, we have a presentation by Krishnakant Saboo – PhD Candidate in ECE under the advisement of Prof. Ravishankar Iyer.

Title:
Reinforcement Learning based Disease Progression Model for Alzheimer’s Disease

Abstract:
Early diagnosis of Alzheimer’s disease (AD) is crucial. However, long term longitudinal data spanning the entire duration of the disease is rarely available. In this talk, I present a model of AD progression that combines biologically motivated differential equations (DEs) and reinforcement learning (RL) with domain knowledge. The model requires limited data for training since it leverages DEs that provide relationships between some, but not all, factors relevant to AD, and utilizes RL to extract the missing relationships. The model predicted individualized 10-year future AD progression better than state-of-the-art learning-based models and provided insights into disease related processes. Finally, I discuss broader applicability of our framework that combines DEs with RL in modelling and understanding other neurological disorders.

Bio:
Krishnakant Saboo is a PhD candidate in the ECE Department at the University of Illinois at Urbana-Champaign advised by Prof. Ravishankar Iyer. He received a Bachelor’s and Master’s degree in Electrical Engineering from the Indian Institute of Technology, Bombay in 2016. His research interests are at the intersection of machine learning, neurology, and neuroscience. For his PhD, he uses ML techniques to model the progression of Alzheimer’s disease and improve the diagnosis of epilepsy. Krishnakant’s work has been recognized by the Mayo/Illinois fellowship, Rambus fellowship, the Elsa and Floyd Dunn award, and the Mavis Future Faculty fellowship. Apart from work, he enjoys teaching, learning about science, and wondering about the meaning of life.

October 15th – Verification of Cyber-Physical Systems

Please join us for this week’s talk by Hussein Sibai-PhD Candidate in ECE under the advisement of Prof. Sayan Mitra:

“Accelerating Verification of Cyber-Physical Systems using Symmetry”.

Abstract:

Scalability is the main challenge facing the verification techniques for cyber-physical systems. Existing methods overlook structural properties of these systems which constrain their behavior and consequently allow, if utilized, for their faster verification. In this talk, I show how key structural information,  such as symmetry, about these systems’ continuous dynamics, can be used to cache computed reachable sets of states and abstract and refine their models to accelerate their safety verification. I present experimental results showing significant speedup in computation time when verifying autonomous drones and cars, even those with neural networks-based controllers, following predefined paths in complex environments.

Bio:

Hussein Sibai is a PhD candidate in the ECE Department at the University of Illinois at Urbana-Champaign advised by Prof. Sayan Mitra. He received a bachelor’s degree in Computer and Communication Engineering from the American University of Beirut in 2014, and a master’s degree in Electrical and Computer Engineering from Illinois in 2017. His research interests are in formal methods, machine learning, and control theory. Hussein has won the best poster award in HSCC 2018, and best paper nominations at HSCC 2017 and ATVA 2019. His work has been recognized by the Rambus fellowship, the Ernest A. Reid fellowship, the MAVIS Future Faculty fellowship, and the ACM SIGBED gold medal for the graduate category in the student research competition in CPS Week’21.

October 8th, 2021 – Impedance Aliasing

This week, we have a presentation by Hyungjoo Seo on impedance aliasing issues in circuits.

Please join us, and stay afterwards to socialize with the other attendees (and eat cookies)!

On impedance aliasing issue in periodically commutating N-path structures — Abstract
The talk will be mainly about challenges posed by impedance aliasing issue in linear periodically time-varying (LPTV) N-path circuits followed by generic LTI-responses, especially for an application of programmable frequency-translational RF front-ends.

Speaker Bio
Hyungjoo Seo received the B.S. degree in EE from the Tokyo Institute of Technology, Tokyo in 2016, and the M.S. degree in ECE from the University of Michigan in 2018. He is currently pursuing the Ph.D. degree in ECE with the University of Illinois at Urbana-Champaign. His current research interests include blocker-tolerant programmable receivers and interference management algorithms for wireless and acoustic applications.

October 1st, 2021 – Imperfect AI Autograders

This week, we have a joint presentation by Tiffany Wenting Li and Silas Hsu on their recently published work regarding imperfect AI autograders.

Please join us, and stay afterwards to socialize with the other attendees (and eat cookies)!

Attitudes Surrounding an Imperfect AI Autograder — Abstract

Deployment of AI assessment tools in education is widespread, but work on students’ interactions and attitudes towards imperfect autograders is comparatively lacking.  This paper presents students’ perceptions surrounding a 90% accurate automated short-answer grader that determined homework and exam credit in a college-level computer science course.  Using surveys and interviews, we investigated students’ knowledge about the autograder and their attitudes.

We observed that misalignment between folk theories about how the autograder worked and how it actually worked could lead to suboptimal answer construction strategies. Students overestimated the autograder’s probability of marking correct answers as wrong, and estimates of this probability were associated with dissatisfaction and perceptions of unfairness.  Many participants expressed a need for additional instruction on how to cater to the autograder. From these findings, we propose guidelines for incorporating imperfect short answer autograders into the classroom in a manner that is considerate of students’ needs.

Personal Bios

Tiffany Wenting Li is a 4th year Ph.D. student in the Computer Science Department at UIUC, advised by Dr. Karrie Karahalios and Dr. Hari Sundaram. Her research interests broadly lie in the intersection of human-computer interaction (HCI), education technology, and artificial intelligence (AI). She is excited about leveraging AI effectively and fairly to increase access to quality education. Currently, she focuses on two lines of research. First, she is working to address the imperfection and opacity of AI-driven feedback systems to maximize students’ learning gain. Her second line of research develops algorithmic systems to facilitate collaborative peer feedback exchanges at scale, optimizing for learning gain, feedback diversity, and efficiency. Before she was a Ph.D. student, she studied mathematics and economics at Cornell University.

Silas Hsu is a 4th year Ph.D. student at UIUC specializing in human-computer interaction (HCI) and advised by Dr. Karrie Karahalios. His research focuses on helping people make the most of imperfect AI systems and helping users correct AI’s mistakes. Currently, Silas pursues this goal in the area of (1) everyday online algorithms that curate ads and feeds, making sure people have meaningful control over the content they see; and (2) algorithms that grade students’ work, giving students tools to help them make sense of possibly imperfect feedback. When he’s not busy, Silas continues to add to his over 20 years of classical piano experience and strives for performance quality rivaling that of professionals.

Sept, 24, 2021 – Pulkit Katare

This week, Pulkit Katare who is part of the Human Centered Autonomy Lab will present!

Please join us, and stay afterwards to socialize with the other attendees (and eat cookies)!

Pulkit Katdare is a fourth year Ph.D Student in the Human Centered Autonomy Lab under Prof. Katherine Driggs-Campbell. In this talk, he is going to be talking about their recent submission to ICRA 2022 titled Off Environment Evaluation Using Convex Risk Minimization. While using robot learning, researchers often use simulators to train the robot and then deploy them on the real world. Robots trained in this manner tend to perform suboptimally in the real world. The reason behind this mismatch is because the simulators don’t generally capture un-certainties of the real world robot, a fact commonly referred to as the Sim2Real gap. In his work, he attempts to bridge this Sim2Real gap by estimating the difference between the simulator and the real world and uses it to augment the RL cost function to be much more reflective of the real world robots.

Let’s Try That Again, Sept. 17

Apologies for the confusion last week. Due to a number of issues, last week’s presentation has been rescheduled to September 17th.

The location of the social hours has also been changed to the conference room, CSL 301, until further notice to accommodate the hybrid Zoom format.  Check your email for the link!

 

This week, Mark Hart will present about the Engineering IT Shared Services and the overall research support landscape here at Illinois.

Come learn about many of the useful resources available to researchers in CSL, and stay afterwards to socialize with the other attendees!

BIO: Mark Hart has been an IT Professional at Illinois since 2004. In his career, he has worked in areas ranging from healthcare to network security, and this has led him to his current role as Assistant Director for Research Services with Engineering IT Shared Services. In his previous capacity as a Research Technology Facilitator, Mark served as an advocate for faculty and researchers, connecting them to the resources they need, and helping ensure those researchers have a voice in shaping the services they use. As AD for Research Services, Mark plans to continue that work with an eye on improving service offerings and support for the entire research workforce within The Grainger College of Engineering.

 

Recording available below:

https://illinois.zoom.us/rec/share/zZBB2uSYKVqkbCTWCrbSdFM3rCTJw0CF7OGhHLS2TkeC6GZa3EbVsfL7XgTuHyAS.00DAsccyspj009xF

CSL Social Hour, Sept. 10, 2021

This week, Mark Hart will present about the Engineering IT Shared Services and the overall research support landscape here at Illinois.

Come learn about many of the useful resources available to researchers in CSL, and stay afterwards to socialize with the other attendees!

BIO: Mark Hart has been an IT Professional at Illinois since 2004. In his career, he has worked in areas ranging from healthcare to network security, and this has led him to his current role as Assistant Director for Research Services with Engineering IT Shared Services. In his previous capacity as a Research Technology Facilitator, Mark served as an advocate for faculty and researchers, connecting them to the resources they need, and helping ensure those researchers have a voice in shaping the services they use. As AD for Research Services, Mark plans to continue that work with an eye on improving service offerings and support for the entire research workforce within The Grainger College of Engineering.

CSL Social Hours Are Back!

After a break caused by COVID-19, the CSL social hours have returned, Fridays at 3pm on the 3rd floor of CSL.

Stop by each week for a short presentation, and stick around afterwards to interact with the broader CSL community!

This website will once again be used to announce speakers, as well as post useful information such as slides.

The first social hour presentation of the academic year was given by Klara Nahrstedt, Director of CSL, and included an overview of the laboratory.
Additionally, the 2021 InstaRecon Innovation Scholarship was awarded to Gizem Tabak during the event. Congratulations!

The slides from the first social hour are below.

Social Hour PPT-9-3-21-final