Past events

The computer science colloquium takes place on Mondays from 11:15 a.m. - 12:15 p.m.

This week's speaker, Catherine Qi Zhao (University of Minnesota; a member of the data science faculty), will be giving a talk titled "Attention in Vision-based AI Systems".

Abstract

Imagine that you are at a bus stop in a new city. You take a few glimpses around, parse and summarize the information you gather, and decide the next steps. Although intuitive, it implies a highly sophisticated and superior ability to select and parse information. Our research is along this line to develop and utilize machine attention for AI systems. In this talk, I will discuss the challenges and share the recent innovations in data, models, and applications from our research.

I will first talk about attention prediction - the ability of machines to find the most relevant information. I will elaborate on our computational models and experimental methods for attention prediction and explain how they have advanced the state-of-the-art. I will then discuss new approaches that leverage attention in computer vision and language tasks, leading to better interpretability and task performance. I will also present preliminary data suggesting that this approach can help reveal and improve the black-box decision-making process of learning-based AI systems. Finally, I will discuss the applications of our models and data in healthcare. I will give two examples where our work leads to the discovery of neurobehavioral signature in autism patients, as well as cutting-edge brain-machine interface technology that restores the lost motor functions in upper-limb amputee patients.

Biography

Catherine Qi Zhao is an assistant professor in the Department of Computer Science and Engineering at the University of Minnesota. Dr. Zhao’s research interests are in computer vision and machine learning, and their applications in healthcare. Her current research on machine attention is supported by NSF and NIH. Dr. Zhao has published more than 80 papers in computer vision and machine learning venues, and other major multidisciplinary journals. Her research has been featured as a cover article at Neuron and Neural Computation, and as oral presentations at CVPR, ICCV, and ECCV. Dr. Zhao is an Area Chair at several premier computer vision conferences and an Associate Editor at TNNLS.

Mark your calendar! The College of Science and Engineering Career Fair is back and going virtual for fall 2020.

The new virtual platform provides features that will allow employers and students to connect personally without waiting in lines.

Tuesday, September 29 and Wednesday, September 30
11 a.m. - 5 p.m. each day

Questions? Contact the CSE Career Center.

Data science seminars hosted by the The Institute for Mathematics and Its Applications (IMA) take place on Tuesdays from 1:25 p.m. - 2:25 p.m.

This week, Vahan Huroyan (University of Arizona), will be giving the lecture.

View the full list of IMA data science seminars.

Mark your calendar! The College of Science and Engineering Career Fair is back and going virtual for fall 2020.

The new virtual platform provides features that will allow employers and students to connect personally without waiting in lines.

Tuesday, September 29 and Wednesday, September 30
11 a.m. - 5 p.m. each day

The computer science and data science program coordinators are available during day 1 of the fair. Learn more about our programs, the admissions process, and the value of a degree in a high demand field. Simply use the Career Fair Plus App to search for the Department of Computer Science & Engineering to schedule an admissions appointment.

Questions? Contact the CSE Career Center.

The computer science colloquium takes place on Mondays from 11:15 a.m. - 12:15 p.m.

This week's speaker, Rui Kuang (University of Minnesota; a member of the data science faculty), will be giving a talk titled "Network-based Machine Learning Methods for Spatial Genomics: A Generalization to High-order Data and Multi-relational Graphs".

Abstract

Biological tissues are composed of different types of structurally organized cell units playing distinct and cooperative functional roles to drive phenotypes such as diseases. The recent spatial transcriptomics technologies have enabled spatially-resolved RNA profiling of single cells mapped with cell identities and localizations for understanding the cells’ organizations and functions. In this talk, I will present a family of network-based machine learning methods that my lab developed to decode the signals in the genomic data for predicting phenotypes. The network-based methods introduce prior information from biological networks and other knowledge graphs for learning with highly structured genomic data. I will explain the biological intuitions, mathematical formulations and algorithms, and experimental results of the network-based learning methods with a focus on how the network-based modeling can be generalized to high-order tensor structures in the new spatial transcriptomics data guided by a multi-relational graph to encode cell spatial information and gene functional information as prior information. I will also provide our perspective on the importance of modeling high-order structures for analyzing spatially-resolved transcriptomes and biological networks and discuss our future plan on developing such high-order learning methods for broader applications in bioinformatics.

Biography

Dr. Rui Kuang is an associate professor in computer science and engineering at the University of Minnesota Twin Cities. His lab is interested in developing machine learning models and algorithms for phenome-genome association analysis by mining knowledge graphs, and phenotype prediction and biomarker identification from gene expression profiling data using network-guided machine learning methods. His lab developed high-order relational learning and meta-analysis methods for integrative studies of multiple knowledge graphs, and single-cell and spatially resolved transcriptomic data. Dr. Kuang is a recipient of NSF CAREER Award in 2011. He received his PhD from Columbia University in 2006, MS from Temple University in 2002 and BS from Nankai University in 1999, all in computer science.

Prospective students can RSVP for an information session to learn about the following graduate programs:

• Computer Science M.S.
• Computer Science MCS
• Computer Science Ph.D.
• Data Science M.S.
• Data Science Post-Baccalaureate Certificate

During the information session, we will go over the following:

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

Data science seminars hosted by the The Institute for Mathematics and Its Applications (IMA) take place on Tuesdays from 1:25 p.m. - 2:25 p.m.

This week, William Leeb (University of Minnesota), will be giving the lecture, "Matrix Denoising with Weighted Loss"

Registration is required to access the Zoom webinar.

Abstract

This talk will describe a new class of methods for estimating a low-rank matrix from a noisy observed matrix, where the error is measured by a type of weighted loss function. Such loss functions arise naturally in a variety of problems, such as submatrix denoising, filtering heteroscedastic noise, and estimation with missing data. We introduce a family of spectral denoisers, which preserve the left and right singular subspaces of the observed matrix. Using new asymptotic results on the spiked covariance model in high dimensions, we derive the optimal spectral denoiser for weighted loss. We demonstrate the behavior of our method through numerical simulations.

Biography

William Leeb is an Assistant Professor in the School of Mathematics at the University of Minnesota, Twin Cities. He earned his PhD from Yale University in 2015 under the supervision of Ronald Coifman, and from 2015 to 2018 was a postdoc in Amit Singer's research group at Princeton University. William's research is in applied and computational harmonic analysis, statistical signal processing, and machine learning. He is particularly interested in estimation problems with low signal-to-noise ratios, high dimensionality, and many nuisance parameters.

View the full list of IMA data science seminars.

The computer science colloquium takes place on Mondays from 11:15 a.m. - 12:15 p.m.

This week's speaker, Sisi Ma (University of Minnesota; a member of the data science faculty), will be giving a talk titled "Multiple Predictively Equivalent Risk Models for Handling Missing Data at Time of Predictions".

Abstract

The presence of missing data at the time of prediction limits the application of risk models in clinical and research settings. Common ways of handling missing data at the time of prediction include measuring the missing value and employing statistical methods. Measuring missing value incurs additional cost, whereas previously reported statistical methods results in reduced performance compared to when all variables are measured. To tackle these challenges, we introduce a new strategy, the MMTOP algorithm (Multiple models for Missing values at Time Of Prediction), which does not require measuring additional data elements or data imputation. Specifically, at model construction time, the MMTOP constructs multiple predictively equivalent risk models utilizing different risk factor sets. The collection of models are stored and to be queried at prediction time. To predict an individual’s risk in the presence of incomplete data, the MMTOP selects the risk model based on measurement availability for that individual from the collection of predictively equivalent models and makes the risk prediction with the selected model. We illustrate the MMTOP with severe hypoglycemia (SH) risk prediction based on data from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study. We identified 77 predictively equivalent models for SH with cross-validated c-index of 0.77 ± 0.03. These models are based on 77 distinct risk factor sets containing 12–17 risk factors. In terms of handling missing data at the time of prediction, the MMTOP outperforms all four tested competitor methods and maintains consistent performance as the number of missing variables increases.

Biography

Dr. Ma is an assistant professor of Medicine in the Division of General Internal Medicine at the University of Minnesota. Dr. Ma's primary research interest is the application of statistical modeling, machine learning, and causal analysis methods in the field of biology and medicine. The questions she seeks answers to include: how to leverage big data and analytical approaches to (1) diagnose and prognose disease and disorders earlier and more accurately. (2) systematically and efficiently identify potential treatment targets for a given disease. (3) identify the best treatment for a particular patient. Dr. Ma also works on theoretical aspects of predictive modeling and causal modeling.

Data science seminars hosted by the The Institute for Mathematics and Its Applications (IMA) take place on Tuesdays from 1:25 p.m. - 2:25 p.m.

This week's speaker, Ju Sun (University of Minnesota), will be giving a talk titled "Does Deep Learning Solve the Phase Retrieval Problem?".

Registration is required to access the Zoom webinar.

Abstract

Phase retrieval is a difficult inverse problem, with three types of intrinsic symmetries. These symmetries can fail even the classic methods, e.g., HIO on complex-valued images when the object support is not precisely known --- allowing free translations. Do these symmetries cause learning difficulty if one deploys the deep learning approach? We show that that’s indeed the case, and we present two solutions to the problem: one is active symmetry breaking based on careful pre-processing on the training data, and the other passive symmetry breaking exploiting massive amounts of data and implicit regularization.

View the full list of IMA data science seminars.

Prospective students can RSVP for an information session to learn about the following graduate programs:

• Computer Science M.S.
• Computer Science MCS
• Computer Science Ph.D.
• Data Science M.S.
• Data Science Post-Baccalaureate Certificate

During the information session, we will go over the following:

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