TBA Speaker's short bio: Hongjiang is a postdoctoral fellow in the Department of Mathematics and Statistics at Auburn University. He obtained his PhD in mathematics at the University of Connecticut in December 2023, under the supervision of Prof. George Yin. Before that, he received his B.S. in Mathematics and Applied Mathematics at Huazhong University of Science and Technology in 2018. Please visit his website https://hongjiang-qian.github.io/index.html (https://hongjiang-qian.github.io/index.html) for more information.

Do you have questions about the new Ultra Course View in HuskyCT? This session will be dedicated to specific questions participants have during the Fall 2025 semester. There will be no set agenda or presentation for this session; EdTech staff will cover any topics participants request. Registrants are encouraged to submit questions ahead of time to edtech@uconn.edu so that staff can prepare detailed answers and examples as necessary.   Register - https://fins.uconn.edu/secure_inst/workshops/workshop_view.php?ser=3652 (https://fins.uconn.edu/secure_inst/workshops/workshop_view.php?ser=3652)

On Wednesday, November 12, 2025, Dr. Andrew Hardaway will be here at UConn Physiology and Neurobiology from UMass Amherst, hosted by Dr. Natale Sciolino. His seminar will be titled, "The central amygdala integrates exogenous and endogenous GLP-1 signals."

Homomorphisms of maximal Cohen-Macaulay modules over the cone of an elliptic curve by Bhargavi Parthasarathy (Syracuse University) Abstract: Consider the ring \(R=k[[x,y,z]]/(f)\) where \(f=x^3+y^3+z^3\) with an algebraically closed field \(k\) and \(char(k) eq 3\). In a 2002 paper, Laza, Popescu and Pfister used Atiyah's classification of vector bundles over elliptic curves to obtain a description of the maximal Cohen-Macaulay modules (MCM) over \(R\). In particular, the matrix factorizations corresponding to rank one MCMs can be described using points in \(V(f)\). If \(M,\ N\) are rank one MCMs over \(R\), then so is \({m Hom}_R(M,N)\). In this talk, I will discuss how the elliptic group law on \(f\) can be used to obtain the point in \(V(f)\) that describes the matrix factorization corresponding to \({m Hom}_R(M,N)\).

Abstract:  Accurate and efficient modeling of urban mobility, along with the prediction of vehicle and human trajectories and object detection in traffic, are crucial for ensuring the safety and resilience of intelligent transportation systems and smart cities. However, the dynamic nature of external environments (such as weather conditions, road networks, neighboring vehicles, and passenger behaviors) or intrinsic changes in target data can lead to significant shifts in data distribution. These shifts can invalidate the trained deep learning models and foster overconfidence in model outputs. In this thesis, we introduce both the learning-based and statistical-based methods to address these issues.First, we design an attention based method for the citywide anomaly event prediction. This method effectively models the spatio-temporal characteristics of urban anomaly events and quantifies the varying impacts of urban mobility on the occurrence of anomaly events. Second, we present an extreme-aware framework to predict the citywide urban mobility under anomalous situations. The proposed framework decomposes the regional urban mobility into spatio-temporally varying regular and extreme dynamics. It minimizes the citywide urban mobility prediction loss under distribution shift. Third, we introduce a conformal prediction based and Gaussian process regression based framework to quantify the output uncertainty of existing trajectory prediction models (base models). This framework aims to improve the prediction accuracy of the base models and reduce the uncertainty of the predicted trajectories under distribution shift. Finally, we introduce an uncertainty-aware adversarial training framework that enhance the resiliency of existing collaborative object detection models for autonomous driving against adversarial attacks. More specifically, this framework alleviates the impacts of adversarial attacks by providing output uncertainty estimation through learning-based module and conformal prediction-based calibration.