Past events

In collaboration with the Minnesota Center for Industrial Mathematics, the Industrial Problems Seminars are a forum for industrial researchers to offer a first-hand glimpse into industrial research. The seminars take place Fridays from 1:25 p.m. - 2:25 p.m.

This week's speaker, Jiguang Shen (Microsoft Research), will be giving a talk titled "Challenges in Building Intelligent Search Systems."

Registration is required to access the Zoom webinar.

Abstract

Intelligent search, powered by natural language processing (NLP) algorithms, helps individuals and enterprise customers find useful information they need at an unprecedented scale. Compared to the traditional web search engines, there are a lot of new challenges in this rising popular domain. In this talk, I will talk about my experience working at the public web search engine Microsoft Bing and the latest work we have done at Microsoft Research & Incubation on building intelligent search systems over enterprise documents.

Biography

Jiguang Shen received his Ph.D. in Applied Mathematics from University of Minnesota in 2017, under the supervision of Professor Bernardo Cockburn. He is currently a Senior Applied Science Manager at Microsoft working on building search and ranking systems.

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 Thursday from 12 p.m. - 1 p.m. during the Fall 2021 semester.

This week's speaker, Irfan Bulu (UnitedHealth Group), will be giving a talk titled "How to improve healthcare AI? Incorporating multimodal data and domain knowledge."

Abstract

Healthcare data is special. Its complex nature is a double-edged sword, possessing great potential but also presenting many difficulties to overcome. For example, administrative claims data —in contrast to the common data types (text, vision, audio) where AI has made eye-popping advances—is multi-modal (i.e., consisting of distinct data types including medical claims, pharmacy claims, and lab results), asynchronous (medication histories and diagnosis histories need not be aligned in time), and irregularly sampled (we only collect data when an individual interacts with the system). Along with such rich and complex data, there is a great deal of domain knowledge in various forms in the healthcare field. In this talk, I will present our work on deep learning architectures for incorporating multimodal data and domain knowledge into models.

Biography

I received a Ph.D. in Physics from Bilkent University in 2007. The focus of my Ph.D. work was novel structures such as photonic crystals, plasmonic devices, and metamaterials for controlling the flow of light. I joined Prof. Marko Loncar’s lab at Harvard University for postdoc after completing Ph.D. There, I tackled problems and challenges in communication security and communication bandwidth using diamond nano-photonic structures. In 2013, I took a career in industrial research at Schlumberger, the largest oil field services company, which started an exciting journey for me in taking innovations from lab to products at the hands of customers. For example, our team invented a new nuclear magnetic resonance logging tool, which improved logging speed by an order of magnitude, thereby addressing an important challenge for our customers in adopting nuclear magnetic resonance measurements. This work also led me to a career in machine learning as both the design of the instrument and interpretation of various measurements in oil field benefited from advances in deep learning. I joined United Health Group in 2018, where I research machine learning algorithms for healthcare applications.

In collaboration with the Minnesota Center for Industrial Mathematics, the Industrial Problems Seminars are a forum for industrial researchers to offer a first-hand glimpse into industrial research. The seminars take place Fridays from 1:25 p.m. - 2:25 p.m.

This week's speaker, Jimmy Broomfield (Target Corporation), will be giving a talk titled "Predicting Tomorrow: Industrial Forecasting at Scale."

You may attend the talk either in person in Walter 402 or registering via Zoom.

Abstract

Have you ever wondered how supply chains make decisions about purchasing and transport? Or perhaps you've stayed up at night wondering how energy companies plan for customer demand. Time series forecasting is a major component used to help business teams solve these problems. In this talk, I'll share my career journey in the world of industrial forecasting. We'll touch on the topics of data preparation, time series models, accuracy metrics, high level architecture, and compute/time constraints.

Biography

Jimmy graduated from the University of Minnesota in 2019 with a PhD in Math and joined Ecolab's advanced analytics team where he primarily worked in the field of time series analysis and forecasting. During his time at Ecolab, Jimmy made contributions to the enterprise's time series classification framework by introducing novel wavelet and frequency based features. He also served as a team lead with the responsibility for architecting, building, and validating a modern supply chain forecasting system for Ecolab's industrial chemical distribution. Jimmy recently made a career transition to the demand forecasting team at Target where he hopes to continue his journey toward understanding industrial forecasting challenges and solutions.

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 Thursday from 12 p.m. - 1 p.m. during the Fall 2021 semester.

This week's speaker, Song Mei (University of California, Berkeley), will be giving a talk titled "The efficiency of kernel methods on structured datasets."

Abstract

Inspired by the proposal of tangent kernels of neural networks (NNs), a recent research line aims to design kernels with a better generalization performance on standard datasets. Indeed, a few recent works showed that certain kernel machines perform as well as NNs on certain datasets, despite their separations in specific cases implied by theoretical results. Furthermore, it was shown that the induced kernels of convolutional neural networks perform much better than any former handcrafted kernels. These empirical results pose a theoretical challenge to understanding the performance gaps in kernel machines and NNs in different scenarios.

In this talk, we show that data structures play an essential role in inducing these performance gaps. We consider a few natural data structures, and study their effects on the performance of these learning methods. Based on a fine-grained high dimensional asymptotics framework of analyzing random features models and kernel machines, we show the following: 1) If the feature vectors are nearly isotropic, kernel methods suffer from the curse of dimensionality, while NNs can overcome it by learning the best low-dimensional representation; 2) If the feature vectors display the same low-dimensional structure as the target function (the spiked covariates model), this curse of dimensionality becomes milder, and the performance gap between kernel methods and NNs become smaller; 3) On datasets that display some invariance structure (e.g., image dataset), there is a quantitative performance gain of using invariant kernels (e.g., convolutional kernels) over inner product kernels. Beyond explaining the performance gaps, these theoretical results can further provide some intuitions towards designing kernel methods with better performance.

Biography

Song Mei is an Assistant Professor in statistics at UC Berkeley. His research is motivated by data science and lies at the intersection of statistics, machine learning, information theory, and computer science. His work often builds on insights that originate within statistical physics literature. His recent research interests include theory of deep learning, high dimensional geometry, approximate Bayesian inferences, and applied random matrix theory.

The Institute for Mathematics and Its Applications (IMA) Data Science Seminars are a forum for data scientists of IMA academic and industrial partners to discuss and learn about recent developments in the broad area of data science. The seminars take place on Tuesdays from 1:25 p.m. - 2:25 p.m.

This week's speaker, Ryan Murray (North Carolina State University), will be giving a talk titled "Data depths meet Hamilton-Jacobi equations."

You may attend the talk either in person in Walter 402 or registering via Zoom.

Abstract

Widespread application of modern machine learning has increased the need for robust statistical algorithms. One fundamental geometric quantity in robust statistics is known as a data depth, which generalizes the notion of quantiles and medians to multiple dimensions. This talk will discuss recent work (in collaboration with Martin Molina-Fructuoso) which connects certain types of data depths with Hamilton-Jacobi equations, a first-order partial differential equation that is fundamental to control theory. Computational considerations, connections to convex geometry and a number of related open problems will all be discussed.

Biography

Ryan Murray received his Ph.D. in mathematics from Carnegie Mellon University in 2016, and was a Chowla Assistant Professor at Penn State University from 2016-2019. Since 2019 he is an assistant professor at North Carolina State University, department of mathematics.

In collaboration with the Minnesota Center for Industrial Mathematics, the Industrial Problems Seminars are a forum for industrial researchers to offer a first-hand glimpse into industrial research. The seminars take place Fridays from 1:25 p.m. - 2:25 p.m.

This week's speaker, Sharath Dhamodaran (OptumLabs), will be giving a talk titled "Data Scientists under attack!! Let's help them together."

You may attend the talk either in person in Walter 402 or registering via Zoom.

Abstract

Imposter Syndrome intensifies each day with growing expectations of being a data scientist. You need to have strong quantitative and technical skills (mathematics, statistics, computer science, operations research, optimization, machine learning), business knowledge and consulting skills (problem formulation and framing), relationship and communication skills (advising, negotiating, and managing expectations), computing skills (general-purpose, statistical, mathematical, databases, business intelligence, big data, cloud), and the list goes on. No one can do it all. Data Scientists that do 70% of these are the best of the best. Let me save you some anxiety by sharing my ongoing journey navigating this challenging and rewarding career.

Biography

I lead a team of data scientists focused on creating healthcare machine learning products for our internal and external customers at OptumLabs, part of UnitedHealth Group. I have 8 years of professional experience solving interesting real-world problems using data science. Outside of work, I enjoy interacting with students and professionals and helping them transition to data science. I also compete in cricket tournaments in Minnesota.

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 Thursday from 12 p.m. - 1 p.m. during the Fall 2021 semester.

This week's speaker, Uday V. Shanbhag (Penn State), will be giving a talk titled "Probability Maximization via Minkowski Functionals: Convex Representations and Tractable Resolution."

Abstract

In this paper, we consider the maximization of a probability P{ζ∣ζ∈K(x)} over a closed and convex set X, a special case of the chance-constrained optimization problem. We define K(x) as K(x)≜{ζ∈K∣c(x,ζ)≥0} where ζ is uniformly distributed on a convex and compact set K and c(x,ζ) is defined as either {c(x,ζ)≜1−|ζTx|m, m≥0} (Setting A) or c(x,ζ)≜Tx−ζ (Setting B). We show that in either setting, P{ζ∣ζ∈K(x)} can be expressed as the expectation of a suitably defined function F(x,ξ) with respect to an appropriately defined Gaussian density (or its variant), i.e. Ep~[F(x,ξ)]. We then develop a convex representation of the original problem requiring the minimization of g(E[F(x,ξ)]) over X where g is an appropriately defined smooth convex function. Traditional stochastic approximation schemes cannot contend with the minimization of g(E[F(⋅,ξ)]) over X, since conditionally unbiased sampled gradients are unavailable. We then develop a regularized variance-reduced stochastic approximation ({\textbf{r-VRSA}}) scheme that obviates the need for such unbiasedness by combining iterative {regularization} with variance-reduction. Notably, ({\textbf{r-VRSA}}) is characterized by both almost-sure convergence guarantees, a convergence rate of O(1/k1/2−a) in expected sub-optimality where a>0, and a sample complexity of O(1/ϵ6+δ) where δ>0.

Biography

Uday V. Shanbhag has held the Gary and Sheila Bello Chaired professorship in Ind. & Manuf. Engr. at Penn State University (PSU) since Nov. 2017 and has been at PSU since Fall 2012, prior to which he was at the University of Illinois at Urbana-Champaign (between 2006–2012, both as an assistant and a tenured associate professor). His interests lie in the analysis and solution of optimization problems, variational inequality problems, and noncooperative games complicated by nonsmoothness and uncertainty. He holds undergraduate and Master’s degrees from IIT,Mumbai (1993) and MIT, Cambridge (1998) respectively and a Ph.D. in management science and engineering (Operations Research) from Stanford University (2006).

The Institute for Mathematics and Its Applications (IMA) Data Science Seminars are a forum for data scientists of IMA academic and industrial partners to discuss and learn about recent developments in the broad area of data science. The seminars take place on Tuesdays from 1:25 p.m. - 2:25 p.m.

This week's speaker, Jie Ding (University of Minnesota), will be giving a talk titled "Organizational Collaboration with Assisted Learning."

You may attend the talk either in person in Walter 402 or registering via Zoom.

Abstract

Humans develop knowledge from individual studies and joint discussions with peers, even though each individual observes and thinks differently. Likewise, in many emerging application domains, collaborations among organizations or intelligent agents of heterogeneous nature (e.g., different institutes, commercial companies, and autonomous agents) are often essential to resolving challenging problems that are otherwise impossible to be dealt with by a single organization. However, to avoid leaking useful and possibly proprietary information, an organization typically enforces stringent security measures, significantly limiting such collaboration. This talk will introduce a new research direction named Assisted Learning that aims to enable organizations to assist each other in a decentralized, personalized, and private manner.

Biography

Jie Ding is an Assistant Professor in Statistics and a graduate faculty in ECE at the University of Minnesota. Before joining the University of Minnesota in 2018, he received a Ph.D. in Engineering Sciences in 2017 from Harvard University and worked as a post-doctoral fellow at Information Initiative at Duke University. Before that, Jie graduated from Tsinghua University in 2012, enrolled in the Math & Physics program and the Electrical Engineering program. Jie has broad research interests in machine learning, with a recent focus on collaborative learning and privacy.

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

In collaboration with the Minnesota Center for Industrial Mathematics, the Industrial Problems Seminars are a forum for industrial researchers to offer a first-hand glimpse into industrial research. The seminars take place Fridays from 1:25 p.m. - 2:25 p.m.

This week's speaker, Juan Restrepo (Oak Ridge National Laboratory), will be giving a talk titled "Research and Opportunities in the Mathematical Sciences at Oak Ridge National Laboratory."

Registration is required to access the Zoom webinar.

Abstract

I will present a general overview of Oak Ridge National Laboratory research in mathematics and computing. A brief description of my own initiatives and research will be covered as well. I will also describe opportunities for students, postdocs, and professional mathematicians.

Biography

Dr. Juan M. Restrepo is a Distinguished Member of the R&D Staff at Oak Ridge National Laboratory. Restrepo is a fellow of SIAM and APS. He holds professorships at U. Tennessee and Oregon State University. Prior to ORNL, he was a professor of mathematics at Oregon State University and at the University of Arizona. He has been a frequent IMA visitor.

His research focuses on data-driven methods for dynamics, statistical mechanics, transport in ocean and uncertainty quantification in climate science.