To start Hilary term, join us in N4.01 on Friday 23rd at 12:30 pm for free pizza and a fun quiz competition. This is the perfect Mathematrix event to come to if you’ve been wanting to swing by for a while and haven’t had the opportunity. 

And see our term card below.

Tue, 10 Feb 2026
12:30
C4

Models for subglacial floods during surface lake drainage events

Harry Stuart
(OCIAM Oxford)
Abstract

As temperatures are increasing, so is the presence of meltwater lakes sitting on the surface of the Greenland Ice Sheet. Such lakes have the possibility of draining through cracks in the ice to the bedrock. Observed discharge rates have found that these lakes can drain at three times the flow rate of Niagara Falls. Current models of subglacial drainage systems are unable to cope with such a large and sudden volume of water. This motivates the idea of a 'subglacial blister' which propagates and slowly dissipates underneath the ice sheet. We present a basic hydrofracture model for understanding this process, before carrying out a number of extensions to observe the effects of turbulence, topography, leak-off and finite ice thickness.

AI assisted triage of UK patients in mental health care services: a qualitative focus group study of patients’ attitudes
Smith, K Hamer-Hunt, J Kormilitzin, A Page, H Joyce, D Cipriani, A BMC Psychiatry volume 26 issue 1 (13 Jan 2026)
Tue, 03 Feb 2026
15:30

Foundations for derived analytic and differential geometry

Kobi Kremnitzer
((Mathematical Institute University of Oxford))
Abstract

In this talk I will describe how bornological spaces give a foundation for derived geometries. This works over any Banach ring allowing to define analytic and differential geometry over the integers. I will discuss applications of this approach such as the representability of certain moduli spaces and Galois actions on the cohomology of differetiable manifolds admitting a \Q-form.

Causal transport on path space
Cont, R Lim, F Annals of Probability
Thu, 26 Feb 2026

12:00 - 12:30
Lecture Room 4, Mathematical Institute

The Implcit Regularisation Effect of Stochastic Gradient Descent for the Ptychography Inverse Problem

Alan Muriithi
((Mathematical Institute University of Oxford))
Abstract
Alan Muriithi will talk about: 'The Implicit Regularisation Effect of Stochastic Gradient Descent for the Ptychography Inverse Problem'
 
Ptychography is a highly non-linear and non-convex inverse problem, typically solved either using a set-projection algorithm or by minimising a variational objective derived from an explicit noise model. While variational formulations more faithfully represent the underlying physics of ptychography, they are often sensitive to initialisation. In practice, methods that minimise a variational objective usually rely on set-projection-based algorithms for their initialisation to obtain visually acceptable reconstructions.
 
This talk investigates the use of stochastic gradient methods in variational ptychographic reconstruction. Viewing stochastic gradient descent (SGD) from a stochastic dynamical systems perspective, we examine how gradient noise influences optimisation in the non-convex regime. In particular, we present evidence suggesting that stochasticity acts as an implicit regulariser for the ptychography inverse problem, improving robustness to poor initialization. Beyond improved robustness, stochastic methods also offer favourable computational scaling for large datasets, which are increasingly prevalent in experimental practice.
Thu, 05 Mar 2026

12:00 - 12:30
Lecture Room 4, Mathematical Institute

TBA

Roy Makhlouf
(UC Louvain)
Abstract

TBA

Thu, 05 Feb 2026

12:00 - 12:30
Lecture Room 4, Mathematical Institute

A Very Short Introduction to Ptychographic Image Reconstruction

Dr Jaroslav Fowkes
((Mathematical Institute University of Oxford))
Abstract

Dr Jari Fowkes will talk about; 'A Very Short Introduction to Ptychographic Image Reconstruction'

 

I will present a very short introduction to the mathematics behind the scientific imaging technique known as ptychography, starting with a brief overview of the physics model and the various simplifications required, before moving on to the main ptychography inverse problem and the three principal classes of optimization algorithms currently being used in practice. 

Tue, 03 Mar 2026
16:00
L6

The hyperbolic lattice point problem

Stephen Lester
Abstract
In this talk I will discuss the hyperbolic circle problem for $SL_2(\mathbb Z)$. Given two points $z, w$ that lie in the hyperbolic upper half‑plane, the problem is to determine the number of $SL_2(\mathbb Z)$ translates of w that lie in the hyperbolic disk centred at z with radius $arcosh(R/2)$ for large $R$. Selberg proved that the error term in this problem is $O(R^{2/3})$. I will describe some recent work in which we improve the error term to $o(R^{2/3})$ as $R$ tends to infinity, for $z,w$ that are CM-points of different, square-free discriminants. This is joint work with Dimitrios Chatzakos, Giacomo Cherubini, and Morten Risager.



 

Subscribe to