Welcome to the Library (1400x200)

Archived Talks and Seminars

Reverse Chronological Order

Algebra/Combinatorics Seminar
Friday, February 28th, 2:00pm, GMCS 405
Speaker Darleen Perez-Lavin, University of Kentucky
Title Color Reconnection Model Meets Quantum Computing
Bio Darleen Perez-Lavin is currently a PhD student in mathematics at the University of Kentucky studying additive and combinatorial number theory under the advisement of Dr. David Leep. She actively works on computing the Plus-Minus Davenport Constant for a direct sum of cyclic groups and finding bounds on the aforementioned object. Her graduate studies are funded by the Department of Defense through the SMART Fellowship. She has also spent time working with the Department of Energy at Fermi Lab outside of Chicago IL. While studying pure mathematics, Darleen looks for opportunities to do applied math. She hopes to be able to contribute to both pure and applied math in the future.
Abstract The color reconnection model is used to explain and predict the production of particles in high energy collisions of hadrons. We can embed this particle physics problem into a graph theory problem where we look for Hamiltonian cycles, similar to the traveling salesman problem. Classically, this question is non-trivial and NP-hard. In this presentation, we demonstrate the limitations of traditional techniques for solving this problem and the possibility of using quantum solvers. In particular, we present an Ising model formulation for quantum annealers. During my time at FermiLab, given by the MSGI-NSF program, I was able to jump in on this problem to help construct an optimal Hamiltonian for quantum annealers. I will be providing an introduction to the physics problem and my contribution in how we used AMPL to help us construct a Hamiltonian.


Mathematics Department Colloquium
Monday, February 24, 4:00pm, Biosciences Center Gold Auditorium
Speaker Andrea Bertozzi, Department of Mathematics, UCLA
Title Graphical Models in Machine Learning, Networks, and Uncertainty Quantification
Bio Andrea Bertozzi is an applied mathematician with expertise in nonlinear partial differential equations and fluid dynamics. She also works in the areas of geometric methods for image processing, crime modeling and analysis, and swarming/cooperative dynamics. Bertozzi completed all her degrees in Mathematics at Princeton. She was on the faculty at Duke University from 1995- 2004 first as Associate Professor of Mathematics and then as Professor of Mathematics and Physics. She has served as the Director of the Center for Nonlinear and Complex Systems while at Duke. Bertozzi moved to UCLA in 2003 as a Professor of Mathematics. Since 2005 she has served as Director of Applied Mathematics. In 2012 she was appointed the Betsy Wood Knapp Chair for Innovation and Creativity. Bertozzi’s honors include the Sloan Research Fellowship, the Presidential Early Career Award for Scientists and Engineers, and SIAM’s Kovalevsky Prize. She was elected to the American Academy of Arts and Sciences and to the Fellows of the SIAM and AMS.
Abstract This talk is an overview of recent work graph models for classification using similarity graphs, for community detection in networks, and for the subgraph isomorphism problem in multichannel networks. The equivalence between the graph mincut problem and total variation minimization on the graph allows one to cast graph-cut variational problems in the language of total variation minimization, thus creating a parallel between low dimensional data science problems in Euclidean space (e.g. image segmentation) and high dimensional clustering. Semi-supervised learning with a small amount of training data can be carried out in this framework with diverse applications ranging from hyperspectral pixel classification to identifying activity in police body worn video. It can also be extended to the context of uncertainty quantification with Gaussian noise models. The problem of community detection in networks also has a graph-cut structure and algorithms are presented for the use of threshold dynamics for modularity optimization. With efficient methods, this allows for the use of network modularity for unsupervised machine learning problems with unknown number of classes. Finally we discuss a different class of graph problem – namely identifying template structure in large world graphs and how combinatorial filtering methods can be structured to efficiently attack this problem with the goal of understanding the entire solution space.


AI Seminar
Thursday, January 28, 2020; 10:30am, GMCS 405
Speaker Dr. Shaozeng Zhang, Department of Anthropology, Oregon State University
Title Contextual interpretation of social media users' interaction with content recommendation algorithm in the “post-truth” era today
Abstract This study examines an online protest movement on location-based social media in order to develop contextual interpretation of user-generated big data and to challenge the ongoing debates on “post-truth”. In 2016, in support of the local protests against a crude oil pipeline passing through the region, social media users all over the world remotely checked in to and wrote location-based reviews of the Standing Rock Indian Reservation, North Dakota using a technique that can be called “location spoofing.” This collective action thus generated a massive volume of “fake” locational information. This study develops an anthropological approach to user-generated big data as digital traces of human activities in the broader social-technological network of the involved human and non-human actors. This study reveals that the online protesters’ use of location-based features and content recommendation algorithm challenged not only the political and technological authorities but also, at a more profound level, the established ways for determining what is true and who gets to decide what is true for what purpose. This unique case of decentralized data generation and dissemination demonstrates an ongoing reconfiguration of the previously established regime of truth that was monopolized by scientists or top politicians. The contextual approach to data interpretation is a pioneering effort to update our epistemological assumptions about truth and our research methodology in data environments today. It calls on scholars across disciplines to envision an epistemological shift in our continuous pursuit of knowledge and truth in the so-called “post-truth” era today.


AI Seminar
Friday, December 6th, 2019; 10:30am, GMCS 405
Speaker Duy H. N. Nguyen, Ph.D., Department of Electrical and Computer Engineering
Title Learning to Estimate and Detect in Communication Systems
Abstract Estimation and detection are among the most primitive statistical signal processing tasks in communication systems. The advancements of communication theory in the past few decades have effectively resolved these tasks with rigorous analysis and predictable performance guarantee, especially in Gaussian channels. This is perhaps the reason why the field of communication theory was reluctant in embracing the booming of machine learning. Nevertheless, there has been a lot of interests and publications in the applications of machine learning for communication systems in the past few years. In this talk, I will revisit some of the basics in estimation and detection for communication systems and examine whether machine learning techniques would be the right tools to perform these tasks, especially in non-Gaussian channels. I will then present some of our recent and ongoing research works in this topic.