**Date/Time**

Date(s) - 20/01/2023*3:30 pm - 4:30 pm*

**Angelica Babei , **Mathematics & Statistics, McMaster University**Title:** * On symmetries of modular forms*

**Abstract:**Modular forms, due to their many symmetries, have become indispensable in studying many phenomena in number theory. For example, they put old results such as the four square theorem in a new light, and have most notably been central to the proof of Fermat’s last theorem. In this short talk, I will present a few ways in which the symmetries of modular forms manifest themselves, with applications to formulas for pi and, if time permits, to properties of the partition function.

**Connor Gregor, **Mathematics & Statistics, McMaster University**Title: ***AutomaticallyClassifying Student Responses with Gradient Boosting Machines***Abstract: **The voice of a single student can easily become lost within the sea of voices that make up an undergraduate math class. The best that one can do as an educator is provide general comments to the entire classroom and hope that its benefit is received by some significant subset of the students. This presentation offers an alternative approach where student voices are organized by means of an automatic classifier. This automatic classifier exists as a gradient boosting machine model that is trained by a set of qualitatively coded student responses which have been formatted into NLP data. Using the classifications suggested by the model, an educator can create tailored recommendations for each category of student voice and improve the quality of teaching received by the class as a whole.**Michel Alexis**, Mathematics & Statistics, McMaster University**Title: ***Two-weight norminequalities for Singular Integral Operators: Testing and Instability***Abstract:** If $T$ is a non-degenerate Calderon-Zygmund operator, like e.g. the Hilbert transform or the Riesz transforms, then it is well-known $T: L^2 (mathbb{R}^n) to L^2 (mathbb{R}^n)$. Whence the weighted problem: when is $T: L^2 (w) to L^2 (w)$, where $w$ is a measure on $mathbb{R}^n$? The answer is still simple: $T$ is bounded if and only if $w$ satisfies the simple $A_2$ Muckenhoupt condition. What about the two-weight problem: when is $T: L^2 (sigma) to L^2 (w)$? Here, no such simple characterization is known. I will discuss recent joint works with Jose Luis Luna-Garcia, Eric Sawyer and Ignacio Uriarte-Tuero which provide a “testing” characterization when the weights sigma^{-1}, w (associated to a more natural rescaled problem) are doubling, and introduce the notion of instability which suggests the “testing” characterization is the best that can be done.

**Location:** Hamilton Hall, Room 305

*Coffee and Cookies will be served in Hamilton Hall Lounge at 3:00 pm. Everyone is welcome.**———————————————————————————————————————————————————————-*

**Dr. Bernhard Banaschewski, **McKay Professor (Emeritus), passed away on October 31, 2022. He was born in Munich on March 22nd, 1926 and became a member of the Department in 1955. Dr. Banaschewski played an influential role in the development of the Department having served as Department Chair twice, from 1961 to 1967 and from 1982 to 1987. The following article from the Hamilton Spectator published on the occasion of the conferring of an honorary degree to Dr. Banaschewski by McMaster University in 2017 offers some details about his interesting and productive life

https://www.thespec.com/news/hamilton-region/2017/06/14/mahoney-do-the-math-dr-banaschewski-adds-another-degree.html

**Dr. Zdzislav Kovarik **passed away on November 29, 2022. Following his Ph.D. at the University of Toronto in 1971, Dr. Kovarik joined the Department in 1973 and worked until his retirement in 2010. An obituary for Dr. Kovarik can be found here

https://www.legacy.com/ca/obituaries/thespec/name/zdislav-kovarik-obituary?pid=203338646