Past events

The UMN Machine Learning Seminar Series brings together faculty, students, and local industrial partners who are interested in the theoretical, computational, and applied aspects of machine learning, to pose problems, exchange ideas, and foster collaborations. The talks are every Wednesday from 11 a.m. - 12 p.m. during the Fall 2022 semester.

This week's speaker, Sijia Liu (Michigan State University), will be giving a talk titled "Robust and Efficient Neural Network Training: A New Bi-level Learning Paradigm".

Abstract

This talk will introduce Bi-level Machine Learning (BML), an emerging but overlooked topic rooted in bi-level optimization (BLO), to tackle the recent neural network training challenges in robust and efficient AI. In the first part, I will revisit adversarial training (AT)–a widely recognized training mechanism to gain adversarial robustness of deep neural networks–from a fresh BML viewpoint. Built upon that, I will introduce a new theoretically-grounded and computationally-efficient robust training algorithm termed Fast Bi-level AT (Fast-BAT), which can defend sign-based projected gradient descent (PGD) attacks without using any gradient sign method or explicit robust regularization.

In the second part, I will move to a sparse learning paradigm that aims at pruning large-scale neural networks for improved generalization and efficiency. As demonstrated by the Lottery Ticket Hypothesis (LTH), iterative magnitude pruning (IMP) is the predominant sparse learning method to successfully find ‘winning’ sparse sub-networks. Yet, the computation cost of IMP grows prohibitively as the sparsity ratio increases. I will show that BML provides a graceful algorithmic foundation for model pruning and helps us close the gap between pruning accuracy and efficiency. Please see the references and codes at Sijia Liu's GitHub repository.

Biography

Dr. Sijia Liu is currently an Assistant Professor at the Department of Computer Science and Engineering, Michigan State University, and an Affiliated Professor at the MIT-IBM Watson AI Lab, IBM Research. His research spans the areas of machine learning, optimization, computer vision, signal processing, and computational biology, with a recent focus on Trustworthy and Scalable ML. He received the Best Paper Runner-Up Award at the Conference on Uncertainty in Artificial Intelligence (UAI) in 2022 and the Best Student Paper Award at the 42nd IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) in  2017. He has published over 60 papers at top-tier ML/AI conferences and (co-)organized several tutorials and workshops on Trustworthy AI and Optimization for Deep Learning at KDD, AAAI, CVPR, and ICML, to name a few.

The John V. Carlis Memorial Lecture is dedicated to the advancement of education and inclusion in the field of computing.

This year's speaker is Dr. Carla Brodley from Northeastern University, giving a talk titled "Broadening Participation in Computing by Opening New Pathways to the BS, MS and PhD.".

Abstract

For the last two decades professors, non-profits, philanthropists, NSF and other agencies have been working to broaden participation in computing (BPC) in higher-ed at all levels and some progress has been made, but often it is incremental and takes place in small pockets. To accelerate the rate of progress in diversifying computing nationally requires that we as a field understand and remove institutional barriers at every level of higher ed, and that we rethink the invitation to create systemic sustainable change.  

Launched in 2019 with funding from Pivotal Ventures LLC, an investment and incubation company created by Melinda French Gates, the Center for Inclusive Computing (CIC) is working in partnership with colleges and universities across the country to increase the representation of women – of all races and ethnicities – in computing. A key focus is to identify and remove institutional barriers and create new pathways to the BS, MS and PhD in computing.  

In her talk, Dr. Carla Brodley, the CIC’s Executive Director and former dean of Northeastern’s Khoury College of Computer Sciences, will explore the most common institutional barriers the CIC is seeing across its portfolio.  She will discuss the concrete measures that can be taken to address barriers to retention and the need to open new pathways to computing, such as creating a BA in computing, making CS1 a general education requirement, handling the distribution of prior computing experience in the intro sequence, creating interdisciplinary BS/BA degrees, rethinking PhD admissions, and creating and scaling the MS in CS for non-computer science majors.

Biography

Professor Carla E. Brodley is the dean of inclusive computing at Northeastern University, where she serves as the executive director for the Center for Inclusive Computing and holds a tenured appointment in Khoury College of Computer Sciences. Brodley served as dean of Khoury College from 2014-2021. Prior to joining Northeastern, she was a professor at the Department of Computer Science and the Clinical and Translational Science Institute at Tufts University (2004-2014). Before joining Tufts, she was on the faculty of the School of Electrical Engineering at Purdue University (1994-2004).

A fellow of the Association for Computing Machinery, the Association for the Advancement of Artificial Intelligence (AAAI), and the American Association for the Advancement of Science (AAAS), Brodley’s interdisciplinary machine learning research led to advances not only in computer science, but in many other fields including remote sensing, neuroscience, digital libraries, astrophysics, content-based image retrieval of medical images, computational biology, chemistry, evidence-based medicine, and predictive medicine. Brodley’s current research focus is computer science education and methods for broadening participation in computing.  In 2022 she was the recipient of the ACM Francis E. Allen Award for Outstanding Mentoring.

Brodley’s numerous leadership positions include serving as program co-chair of the International Conference on Machine Learning, co-chair of AAAI, and associate editor of the Journal of AI Research and the Journal of Machine Learning Research. She previously served on the Defense Science Study Group, the board of the International Machine Learning Society, the AAAI Council, the executive committee of the Northeast Big Data Hub, DARPA’s Information Science and Technology Board, and the NSF CISE advisory committee.  Brodley is currently vice chair of the Board of Trustees of the Jackson Laboratory (JAX).  She serves as a member of the Board of Directors of Alegion, Inc, of the Computing Research Association Board of Directors, of the Mass Technology Leadership Council, of the International Advisory Board of the Quad Fellowships, and is a strategic advisor for Science for America.

RSVP today!.

During each session, the MSSE staff will review:

• Requirements (general)
• Applying
• Prerequisite requirements
• What makes a strong applicant
• Funding
• Resources
• Common questions
• Questions from attendees

The computer science colloquium takes place on Mondays from 11:15 a.m. - 12:15 p.m. This week's speaker, Stephanie Forrest (Arizona State University), will be giving a talk titled "The Biology of Software".

Abstract

Computer programmers like to think of software as the product of intelligent design, carefully crafted to meet well-specified goals.  In reality, large software systems evolve inadvertently through the actions of many individual programmers, often leading to unanticipated consequences.  Because software is subject to constraints similar to those faced by evolving biological systems, we have much to gain by viewing software through the lens of evolutionary biology.  The talk will highlight research applying the mechanisms of evolution quite directly to software, including repairing bugs and runtime optimization of GPU codes.  These results have implications for how we think more generally about engineering complex systems that are subject to evolutionary pressures and engineering constraints.

Biography

Stephanie Forrest is Professor of Computer Science at Arizona State University, where she directs the Biodesign Center for Biocomputation, Security and Society.  Her interdisciplinary research focuses on the intersection of biology and computation, including cybersecurity, software engineering, and biological modeling.

Prior to joining ASU in 2017, she was a Distinguished Professor at the University of New Mexico and served as Dept. Chair. She is a member of the Santa Fe Institute External Faculty and past co-Chair of its Science Board and Interim VP for Academic Affairs.  She spent 2013-2014 as a Jefferson Fellow at the U.S. Dept. of State as a Senior Science Advisor for cyberpolicy.  She was educated at St. John's College (B.A.) and the University of Michigan (M.S. and Ph.D. in Computer Science).

Some of her awards include: The 2020 Test of Time Award from the IEEE Security and Privacy Symposium; The 2019 Most Influential Paper Award from the International Conference on Software Engineering, the Santa Fe Institute Stanislaw Ulam Memorial Lectures (2013), the ACM/AAAI Allen Newell Award (2011), and the NSF Presidential Young Investigator Award (1991).  She is a Fellow of the IEEE and a member of the Computing Research Association Board of Directors.

The Minnesota Natural Language Processing (NLP) Seminar is a venue for faculty, postdocs, students, and anyone else interested in theoretical, computational, and human-centric aspects of natural language processing to exchange ideas and foster collaboration. The talks are every other Friday from 2 - 3 p.m. during the fall 2022 semester.

This week's speaker, Muhao Chen (USC), will be giving a talk titled "Robust and Indirectly Supervised Information Extraction".

Abstract

Information extraction (IE) is the process of automatically inducing structures of concepts and relations described in natural language text. It is the fundamental task to assess the machine’s ability for natural language understanding, as well as the essential step for acquiring structural knowledge representation that is integral to any knowledge-driven AI systems. Despite the importance, obtaining direct supervision for IE tasks is always very difficult, as it requires expert annotators to read through long documents and identify complex structures. Therefore, a robust and accountable IE model has to be achievable with minimal and imperfect supervision. Towards this mission, this talk covers recent advances of machine learning and inference technologies that (i) grant robustness against noise and perturbation, (ii) mitigate spurious correlations, and (iii) provide indirect supervision for logically consistent and label-efficient IE.

Biography

Muhao Chen is an Assistant Research Professor of Computer Science at USC, and the director of the USC Language Understanding and Knowledge Acquisition (LUKA) Lab (https://luka-group.github.io/). His research focuses on robust and minimally supervised machine learning for natural language understanding, structured data processing, and knowledge acquisition from unstructured data. His work has been recognized with an NSF CRII Award, faculty research awards from Cisco and Amazon, and an ACM SigBio Best Student Paper Award. Dr. Chen obtained his Ph.D. degree from UCLA Department of Computer Science in 2019, and was a postdoctoral researcher at UPenn prior to joining USC.

Please join the Department of Computer Science & Engineering (CS&E) for a special celebration of the life of James E. Parker (1984 - 2022), a beloved teacher, mentor and friend.

Enjoy hosted beverages and appetizers, and the chance to connect James' family, friends, classmates, colleagues and students. 

Questions? Contact cscievents@umn.edu.

This is a private reception and RSVPs are required. Please register by October 26. 

 

The UMN Machine Learning Seminar Series brings together faculty, students, and local industrial partners who are interested in the theoretical, computational, and applied aspects of machine learning, to pose problems, exchange ideas, and foster collaborations. The talks are every Wednesday from 11 a.m. - 12 p.m. during the Fall 2022 semester.

This week's speaker, Juggy Jagannathan (3M health), will be giving a talk titled "AI and Future of Healthcare."

Abstract

AI technology is permeating every aspect of healthcare. It is transforming how care is provided in acute, ambulatory and home settings. It is transforming how population health is being managed. AI in combination with other disciplines, such as genetics and virtual reality, is accelerating innovative treatments. Drug discovery and clinical trials are being transformed. This presentation will provide an overview of where AI is already being used and prognosis for the immediate future.

Biography

V. “Juggy” Jagannathan, PhD, researches artificial intelligence and computer science for 3M Health Information Systems, developing new natural language and deep learning technologies that automatically structure patient-physician conversations into clinical documents. As an AI researcher with over four decades in the field, Juggy’s background includes both industry technology development and academia. The company he founded in 1996 with technology from West Virginia University (WVU) was acquired by M*Modal in 2003 and he has been with the company ever since. He is also an adjunct professor teaching Computer Science and Natural Language Processing (NLP) at WVU. Juggy is a tennis nut, golf nut and a yoga nut. In short, a nut.

The second seminar in our 2022 MSSE Industrial Seminar series will feature Ken Kousen, a Java Champion and Grails RockStar. Kousen will be giving a talk titled "Managing Your Manager" as a technical professional. The event is open to all MSSE students, University of Minnesota faculty, alumni, and guests.

Abstract

Conflict between technical professionals and traditional managers is inevitable, because you want different things. Worse, most employees feel that when conflicts arise, their only options are either to go along with what the manager wants, or leave. Neither option gets you what you want when you want it.

This talk discusses a third option: how to build a relationship over time that makes your boss an ally. The goal is to build a productive relationship that allows you to push back against decisions you don't like, while maintaining a constructive, loyalty-based relationship that satisfies both sides. Topics will include the two messages to keep in mind whenever you interact with the boss, how to use solutions to the iterated Prisoner's Dilemma problem to resolve conflicts, how to structure communications in a way most likely to be heard and understood, and more.

Bio

Ken Kousen is a Java Champion, Oracle Groundbreaker Ambassador, and a Grails Rock Star. He is the author of the Pragmatic Library books "Help Your Boss Help You" and "Mockito Made Clear," the O'Reilly Media books "Kotlin Cookbook", "Modern Java Recipes", and "Gradle Recipes for Android", and the Manning book "Making Java Groovy". He also has recorded over a dozen video courses for the O'Reilly Learning Platform, covering topics related to Android, Spring, Java, Groovy, Grails, and Gradle.

In 2013, 2016, and 2017 he won a JavaOne Rockstar award. His academic background includes BS degrees in Mechanical Engineering and Mathematics from M.I.T., an MA and Ph.D. in Aerospace Engineering from Princeton, and an MS in Computer Science from R.P.I. He is currently President of Kousen IT, Inc., based in Connecticut.

 Join the Department of Computer Science & Engineering (CS&E) for this all-alumni event featuring Minnesota spatial computing faculty. Enjoy hosted beverages and appetizers, and the chance to reconnect with former classmates, colleagues, instructors, and friends. All alumni of the University of Minnesota CS&E programs (Computer Science, Data Science, MSSE) are invited to attend, and guests are welcome. 

Questions? Contact cscievents@umn.edu.

About the Program

Learn about current and future directions in spatial computing research from experts in the CS&E community!

• Jayant Gupta, Ph.D. student
  
  The rise of spatial big data (e.g., trajectories, remote-sensing) is fueling growth of Geo-AI to speed-up creation of new maps and discover novel spatiotemporal patterns of our lives. However, generic methods are hampered by the high cost of false errors, spatial variability, etc. We share novel spatial data science methods (e.g., SaTScan co-location miner) and case studies with societal applications. 
   
• Mohamed Mokbel, Professor
  
  Spatial Computing Systems: Embedding spatial awareness boosts scalability, accuracy, and efficiency of computing system infrastructures for big data, knowledge discovery, data cleaning, and machine learning. This presentation will discuss the impact of spatial computing systems in various applications along with UMN SpatialHadoop, a pioneering geospatial cloud computing system.
   
• Yao-Yi Chiang, Associate Professor 
  
  Spatially-enable AI: Spatial concepts and methods enhance the effectiveness of Artificial Intelligence (AI) in many areas including health, transportation, and environmental sciences. This presentation will share spatially-enable AI technologies for air quality predictions, data-driven intelligent transportation systems, human mobility mining, and automatic reading of thousands of historical map scans in archives for understanding human-induced changes in the environment.

There is no charge to attend our event, but pre-registration is required. Register by Tuesday, Oct. 18.