ECM Seminar: Tomorrow’s Mathematicians Today 2023 Winners

Event


Date:

Time : 4:00 pm - 5:00 pm

Online via Zoom

UK

Organiser: Early Career Mathematicians Committee

Organiser Email: ecm@ima.org.uk

Tuesday June 11, 2024 4:00 pm Tuesday June 11, 2024 5:00 pm Europe/London ECM Seminar: Tomorrow’s Mathematicians Today 2023 Winners Online via Zoom, , , , UK In the latest run of the Early Career Mathematicians Seminar Series, we will be joined by the winners of the […] Event Link: https://ima.org.uk/24223/ecm-seminar-tomorrows-mathematicians-today-2023-winners/ Early Career Mathematicians Committee ecm@ima.org.uk

ECM Seminar: Tomorrow’s Mathematicians Today 2023 Winners


In the latest run of the Early Career Mathematicians Seminar Series, we will be joined by the winners of the Tomorrow’s Mathematicians Today 2023 prizes. Maria Esteban Casadevall (Heriot-Watt) won the IMA Prize for the best presentation and Veronica Bitonti (University College London) won the IMA Popular Vote Prize. Maria and Veronica will be speaking for 30 minutes each, see below for their abstracts.

Maria’s talk: Geometric methods for community detection: Discrete notions of curvature in graphs and clustering in the Hyperbolic Space

In recent years, a wide number of geometrical tools have been developed to exploit the intrinsic properties of complex networks. In this talk we will explore two of these notions and understand its application to community detection problems.

The first part of the talk will focus on the Olliver Ricci curvature, a discrete notion of edge curvature based on the Wasserstein distance between probability distributions around neighbouring nodes. We will begin with a definition and description of such notion, together with a justification of its connection to the Ricci curvature in manifolds. A discrete version of the Ricci Flow can then be defined iteratively to flatten the curvature of the graph, bringing important applications to community detection.

The second part of the talk will explore the Hyperbolic space, which has lately been receiving increasing attention due to its success in representing learning for hierarchical data. We will begin with a definition and description of such spaces, together with a justification of why its geometric properties are particularly fit for hierarchical data. We will then discuss how the K-Means Clustering Algorithm can be defined in the Hyperbolic Space and we will see an implementation of such to a real-world network.

Veronica’s talk: The quest for structure: Continued Fractions in Enumerative Combinatorics

Dealing with patterns and how they can be formed is one of the main objects of studies in enumerative combinatorics, and an essential tool to understand such properties is through the majestic method, used already by Euler in 1746, to represent ordinary generating functions as a continued fraction. In particular, in 1980, Flajolet analysed in great detail the connection between Stieltjes-type and Jacobi-type continued fractions and generating functions for Dyck and Motzkin Paths with height-dependent weights. Further developments in this direction has expanded such knowledge to Thron-type continued fractions (T-fractions) which can be interpreted in terms of Schröder Paths. Recently such ideas were extended by Pétréolle, Shu, Sokal in 2018 to what they call “branched continued fractions”. In this presentation, the aim is to introduce the main concepts of the subject by looking at some classes of labelled increasing trees.

Bios:

Maria Esteban Casadevall, Heriot-Watt University: www.linkedin.com/in/mariaestebanc

Maria is a final year Mathematics student at Heriot-Watt university. She is originally from Barcelona, in Spain, where she lived for the first eighteen years of her life, before moving to Scotland for her undergraduate studies. During her degree, she spent a year in exchange at the University of Waterloo, in Canada, and a year in industry working as a Technology Consultant in Financial Services at Ernst&Young, in London. Next year she will be continuing her studies with a Master’s in Mathematical Science at the University of Oxford. Her primary interest lies in the intersection between geometry, topology, and machine learning.

Veronica Bitonti, University College London: http://linkedin.com/in/veronica-bitonti-b0857919a

Veronica is a final year undergraduate student at University College London. Even though she is still at the begging of her academic career, the area she would like to work on is combinatorics and complexity theory. In particular, she has done a reading project on Turing Machines with the result of a presentation, “A Mathematical View of Space: the Universe as a TM”, for the Tomorrow’s Mathematicians Today Conference (TMT) 2022. She then, through a reading project done at UCL with Dr Strouthous, discovered Dehn’s problems and the idea of undecidability in group theory, which was then presented in the talk entitled “Intellectual Challenges: Algorithms and Undecidable problems in Group Theory” at the TMT 2023 Conference. Currently, she is working with Prof. Sokal on a research project on continuous fractions in enumerative combinatorics.

When Veronica is not doing maths, she is most certainly either doing archery or cultivating her passion for the humanities. If she had not read mathematics at university, she would probably have studied literature!

To book

If you would like to attend this online event, please register to receive the zoom link via https://my.ima.org.uk/item.php?eventid=436

If you are an IMA Member or you have previously registered for an IMA conference, then you are already on our database. Please “request a new password” using the email address previously used, to log in.

You will receive your Zoom link for the event 1 day prior.

Published