Fall 2026
Location and Time: Minard 230 at 3:00 PM (Refreshments at 2:30 in Minard 404)
*Special Colloquia or Tri-College Colloquia venues and times may vary, please consult the individual listing.
Tuesday, October 27
Thanh Phat Vo, UND
Coderivative-based Newton methods in structured nonconvex and nonsmooth optimization
Abstract: This talk proposes and develops new Newton-type methods to solve structured nonconvex and nonsmooth optimization problems with justifying their fast local and global convergence by means of advanced tools of variational analysis and generalized differentiation. The objective functions belong to a broad class of prox-regular functions with specification to constrained optimization of nonconvex structured sums. We also develop a novel line search method, which is an extension of the proximal gradient algorithm while allowing us to globalize the proposed coderivative-based Newton methods by incorporating the machinery of forward-backward envelopes. Applications and numerical experiments, which are provided for nonconvex least squares regression models, Student’s t-regression with an -penalty, and image restoration problems, demonstrate the efficiency of the proposed algorithms.
Tuesday, November 3
*Special Tri-College Colloquium at NDSU
Juliann Geraci, Concordia
How Can Algebra Measure Combinatorial Complexity?
Abstract: How can we measure the complexity of a matrix whose entries are only zero or one? Boolean matrix factorization provides one way to answer this question by expressing a binary matrix as a combination of simpler matrices using logical operations. This problem has a natural interpretation in graph theory, where it becomes a question about covering the edges of a bipartite graph with complete bipartite subgraphs.
We will explore how tools from commutative algebra can help us understand this combinatorial problem. By associating monomial ideals to graphs and matrices, we obtain algebraic lower bounds on Boolean rank and uncover connections with other measures of combinatorial complexity. I will also discuss related work on neural codes, mathematical objects used to represent patterns of neural activity and explain how algebraic structures reveal connections between neural coding and Boolean matrix factorization. Throughout, the goal is to illustrate how ideas from algebra, combinatorics, and discrete mathematics can inform one another.
Tuesday, November 10
Maxx Kureczko, NDSU
The Successful Student: Optimizing Learning in Large Lectures
Abstract: Large-lecture Math 144 (Mathematics for Business) at North Dakota State University had a high DFW rate (grades of D, F, or withdrawal) when taught in a traditional format. Over the past eighteen semesters, through the implementation of evidence-based and student-centered changes, the DFW rate has been halved without sacrificing standards or content. We will present our three-fold approach to increasing student success: group work with mandatory attendance, backward design of course content and assessments, and the effective use of SCALE-UP (active learning) classrooms. After a brief overview of the history of Math 144, we will examine student data and measure the effectiveness of these techniques when implemented at NDSU. We will conclude with a discussion of how these methods can be used in classrooms of any size to improve student success and retention.
Tuesday, November 17
Warren Christensen, NDSU
TBD
Abstract: TBD
Tuesday, December 1
*Special Tri-College Colloquium at MSUM
Talk: TBD
Janet Page, NDSU
TBD
Abstract: TBD