Past

Alday & Maldacena conjectured an equivalence between string amplitudes in AdS5 ×S5 and null polygonal Wilson loops together with a duality with amplitudes for planar N = 4 super-Yang-Mills (SYM).  At strong coupling this identifies SYM amplitudes with (regularized) areas of minimal surfaces in AdS.  They reformulated the minimal surface problem as a Hitchin system and in collaboration with Gaiotto, Sever & Vieira they introduced a Y-system and a thermodynamic Bethe ansatze (TBA) expressing the complete integrability that could in principle be used to solve for the amplitude at strong coupling. This lecture will review the parts of this material that we need and use them to identify new geometric structures on the spaces of kinematics for super Yang-Mills amplitudes/null polygonal Wilson loops.   In AdS3, the kinematic space is the cluster variety  M_{0.n} X M_{0,n}, where M_{0,n} is the moduli space of n points on the Riemann sphere moduli Mobius transformations.   The nontrivial part of these amplitudes at strong coupling, the remainder function,  turns out to be the (pseudo-)K ̈ahler scalar for a (pseudo-)hyper-Kaher geometry. It satisfies an integrable system and we give its its Lax form. The result follows from a new perspective on Y-systems more generally as defining the natural twistor space associated to the hyperkahler geometry of the Hitchin moduli space for these minimal surfaces. These connections in particular allows us to prove that  the amplitude at strong coupling satisfies the Plebanski equations for a hyperKahler scalar for  these pseudo-hyperk ̈ahler and related geometries. These hyperkahler geometries are nontrivial, (not semiflat) with a nontrivial TBA that encodes the mutations of the cluster structure.  These new structures underpinning the N=4 SYM amplitudes  will be important beyond strong coupling.  This is based on joint work with Hadleight Frost and Omer Gurdogan, https://arxiv.org/abs/2306.17044.

The representation theory of Lie algebras over fields of positive characteristic behaves quite differently to the characteristic zero case. For example, in positive characteristic, the dimension of all simple modules is finite and bounded. In this talk, we’ll begin by recalling the classification of finite simple representations of sl_2, and then explore how this changes when we move to the positive characteristic setting. Along the way, we’ll discuss the additional structures that appear in positive characteristic, such as restricted Lie algebras, the p-centre, and reduced enveloping algebras.

ACFA is the model companion of the theory of a field endowed with a distinguished endomorphism. This theory has been extensively studied by Chatzidakis and Hrushovski. Notably, it was shown that any non-principal ultraproduct of algebraically closed fields with powers of the Frobenius map gives rise to a model of ACFA.

In this talk, I will discuss the model theory of pairs of ACFA. In particular, we will give an axiomatization of those pairs in which the smaller one is transformally algebraically closed in the larger one. These are precisely the ultraproducts of pairs of algebraically closed fields equipped with powers of the Frobenius map. This theory also provides an example of beautiful pairs in the sense of Cubides Kovacsics, Hils, and Ye.

This is joint work with Martin Hils, Udi Hrushovski, and Jinhe Ye.

Graph causal optimal transport is a recent generalisation of causal optimal transport in which the allowed couplings satisfy causal restrictions given by a directed graph. Inspired by applications to structural causal models, it was originally introduced in Eckstein and Cheridito (2023). We study fundamental properties of graph causal optimal transport, with a particular focus on its induced Wasserstein distance. Our main result is a full characterisation of the directed graphs for which this associated Wasserstein distance is indeed a metric, an open problem in the original paper. We fully characterise the gluing properties of graph causal couplings, prove denseness of Monge maps, and provide a dynamic programming principle. Finally, we present an application to continuity of stochastic team problems. Based on joint work with Jan Obloj.

Over 30 years has passed since the original proof of Fermat's Last Theorem by Wiles and Taylor—Wiles. There are now several proofs known to humanity, and I'm currently teaching one of them to a computer. This made me try to find out what the most ergonomic route was nowadays, and I found it by asking Richard Taylor what it was. In the talk I will summarise how to prove Fermat's Last Theorem in 2026, highlighting the differences between the modern method and the original route discovered by Wiles (we do use p=3, but in a different way). I won't talk much at all about Lean and essentially none of the work I will present is my own; this will just be a standard number theory seminar, and probably everything in it will already be known to the experts, but hopefully younger people will learn something.

Dr Jindong Wang will talk about; 'Stabilised Finite Element Methods for General Convection–Diffusion Equations'

This talk presents several stabilised finite element methods for general convection–diffusion equations, with particular emphasis on recent extensions to vector-valued problems arising in magnetohydrodynamics (MHD). Owing to the non-self-adjoint structure of the operator and the potentially large disparity between convective and diffusive scales, standard Galerkin discretisations may exhibit non-physical oscillations. We design a class of upwind-type schemes and exponentially fitted methods for vector-valued problems that mitigate these effects, highlighting both their shared stabilisation mechanisms and the distinctive features that arise in the vector-valued setting. These developments illustrate concrete strategies for the design and analysis of finite element discretisations for general convection–diffusion problems.

Roy Makhlouf will talk about: 'Random Embeddings for Global Optimization: Convergence Results Beyond Low Effective Dimension'
 

Timely optimization problems are high-dimensional, calling for dimensionality reduction techniques to solve them efficiently. The random embedding strategy, which optimizes the objective along a low-dimensional subspace of the search space, is arguably the simplest possible dimensionality reduction method. Recent works quantify the probability of success of this strategy to solve the original problem by lower bounding the probability of a random subspace to intersect the set of approximate global minimizers. These works showed that, when the objective has low effective dimension (i.e., is only varying along a low-dimensional subspace of the search space), random embeddings of sufficiently large dimension solve the original high-dimensional problem with probability one. In this work, we relax the low effective dimension assumption by considering objectives with anisotropic variability, namely, Lipschitz continuous functions whose Lipschitz constant is small (though nonzero) when the function is restricted to a high-dimensional subspace. Exploiting tools from stochastic geometry, we lower bound the probability for a random subspace to intersect the set of approximate global minimizers of these objectives, hence, the probability of random embeddings to succeed in solving (approximately) the original global optimization problem. Our findings offer deeper insights into the role of the dimension of the optimization problem in this probability of success.

Neurons interact via spikes, which is a pulse-like, discontinuous mechanism. Their mean-field PDE description gives Fokker-Planck equations with novel nonlinearities. From a probability point of view, these give rise to Mckean-Vlasov equations involving hitting times. Similar mechanisms also arise in models for systemic risk in mathematical finance, and the supercooled Stefan problem. In this talk, we will first present models for spiking neurons: both microscopic particle models and macroscopic PDE models, with an emphasis on the general mathematical structure. A central question for these equations is the finite-time blow-up of the firing rate, which scientifically corresponds to the synchronization of a neuronal network. We will discuss how to continue the solution physically after the blow-up, by introducing a new timescale. The new timescale also helps us to understand the long term behavior of the equation, as it reveals a hidden contraction structure in the hyperbolic case. Finally, we will present a recently developed numerical solver based on this framework. Numerical tests show that during the synchronization the standard microscopic solver suffers from a rather demanding time step requirement, while our macro-mesoscopic solver does not.

In this talk I will present a few topics of recent interest that centre around the theme of “driven interfacial hydrodynamics”: fluid mechanical systems in which droplets and particles are self-propelled through interaction with the environment. I will also present some very recent work on using differentiable physics (a branch of physics-informed machine learning) to determine constitutive relations for highly plasticised metals.

This talk will contain elements of fluid dynamics, experimental mechanics, dynamical systems, statistical physics, and machine learning.

This talk will provide an overview of the landscape of bicommutant categories, these are tensor categories with a strong functional-analytic flavour. I will discuss the evolution of the definition (and give the current version of the definition) and explain precisely how they categorify von Neumann algebras, in the same way a tensor category can be viewed as a categorification of an algebra. We will also introduce the string-calculus that renders the coherences in the definition transparent and workable. 

The necessary background from functional analysis (in particular, operator theory) will be reviewed, and I will conclude with open questions (if waiting for the end of talk is not your style, there are 75 Open problems on André’s website). 

A central goal in the study of quantum chaos is being able to make universal statements about the dynamics of generic Hamiltonian systems. Under time evolution, an initially local operator progressively explores the Hilbert space of a system becoming increasingly non-local in the process. We will see that this idea lends itself to a natural notion of operator complexity measured (in the Hilbert space of operators) by the overlap of a time-evolving operator with a basis naturally adapted to time evolution and stratified by the growth in the operator's support. The information contained in this so-called Krylov basis is encoded in a sequence called the Lanczos coefficients which quantify the rate at which an operator is "pushed" along the Krylov basis to successively more complex elements. The universal operator growth hypothesis is then the conjecture that the Lanczos coefficients grow asymptotically linearly in any quantum chaotic system. In this talk, I will present an overview of these ideas and see how they manifest in the example of the well-studied SYK model. This talk is primarily based on 1812.08657.

I will discuss a 2-dimensional model of random walk in random environment known as line model. The environment is described by two independent families of i.i.d. random variables dictating rates of jumps in vertical, respectively horizontal directions, and whose values are constant along vertical, respect. horizontal lines. When jump rates are heavy-tailed in one of the directions, the random walk becomes superdiffusive in that direction, with an explicit scaling limit written as a two-dimensional Brownian motion time-changed (in one of the components) by a process introduced by Kesten and Spitzer in 1979. I will present ideas of the proof of this result, which relies on appropriate time-change arguments.  In the case of a fully degenerate environment, I will present a sufficient condition for non-explosion of the process (which is also believed to be sharp), as well as conjectures on the associated scaling limit.

This is based on joint work with J.-D. Deuschel (TU Berlin). 

(Joint seminar with Random Matrix Theory)

Graph products of groups were introduced by Green as a construction that encompasses both direct products and free products. Likewise, the notion of graph product of von Neumann algebras, introduced by Caspers and Fima, recovers both tensor products and free products. Camille Horbez will present rigidity theorems for graph products of tracial von Neumann algebras, and discuss the computation of their symmetries, drawing parallels with the case of groups. This is a joint work with Adrian Ioana. 

Random walks on graphs can mix slowly. To speed it up, imagine that at each step instead of choosing the neighbor at random, there is a small probability $\varepsilon > 0$ that we can choose it. We show that in this case, at least for graphs of bounded degree, there is a way to steer the walk so that we visit every vertex in $n^{1+o(1)}$ many steps. The key to this result is a way to decompose arbitrary graphs into small-diameter pieces.

I will discuss some recent progress on the Donaldson Segal programme, and in particular how calibrated cycles (coassociative submanifolds, special Lagrangians) arise from the large mass limit of $G_2$ and Calabi Yau monopoles.

Thompson’s groups, introduced by Thompson in 1965, have had a lot of attention in the last fifty years. Being finitely presented, a natural question is to compute their Dehn function. All three groups are conjectured to have quadratic Dehn function; this conjecture was confirmed for Thompson’s group 𝐹 by Guba in 2006. During Matteo Migliorini's talk, we show how to deduce from Guba’s result that Thompson’s group 𝑇 has a quadratic Dehn function as well.

Koszul algebras are positively graded algebras with very amenable homological properties. Typical examples include the polynomial ring over a field or the exterior and symmetric algebras of a vector space. A category is called Koszul if it has a grading with which it is equivalent to the category of graded modules over a Koszul algebra. A famous example (due to Soergel) is the principal block of category $\mathcal{O}$ for a semisimple Lie algebra. Koszulity is a very nice property, but often very difficult to check. In this talk, Thorsten Heidersdorf (Newcastle University) will give a criterion that allows to check Koszulity in case the category is a graded semi-infinite highest weight category (which is a structure that appears often in representation theory). This is joint work with Jonas Nehme and Catharina Stroppel.

Recent years have seen the expansion of the traditional notion of symmetry in quantum theory to so-called generalised or categorical symmetries, which may in particular be non-invertible. This seems to be at odds with Wigner's theorem, which asserts that quantum symmetries ought to be implemented by (anti)unitary -- and hence invertible -- operators on the Hilbert space. In this talk, we will try to resolve this puzzle for generalised symmetries that are described by (higher) fusion categories. After giving a gentle introduction to the latter, we will discuss how one can associate an inner-product-preserving operator to (possibly non-invertible) symmetry defects and illustrate our construction through concrete examples. Based on the recent work 2602.07110 with Gai and Schäfer-Nameki.

In this talk we describe several aspects related to the theory of some anisotropic parabolic equations. The anisotropy shown in such equations will appear in the form of porous medium, in the fast and porous medium diffusion regime. In particular, we show the existence of selfsimilar fundamental solutions, which is uniquely determined by its mass, and the asymptotic behaviour of all finite mass solutions in terms of the family of self-similar fundamental solutions. Time decay rates are derived as well as other properties of the solutions, like quantitative boundedness, positivity and regularity. 

The investigation of both models are objects of joint works with F. Feo and J. L. V´azquez.

Any two 1 by 1 real matrices commute.  This is in general not the case for 2 by 2 real matrices.  However, if A, B, C, and D are any 2 by 2 real matrices, then ABCD - ABDC - ACBD + ACDB + ADBC - ADCB - BACD + BADC + BCAD - BCDA - BDAC + BDCA + CABD - CADB - CBAD + CBDA + CDAB - CDBA - DABC + DACB + DBAC - DBCA - DCAB + DCBA = 0.  This identity is the first instance of a general result of Amitsur and Levitski; I will explain a simple graph-theoretic proof due to Swan.

In the first part of the talk, I will present an overview of recent advances in the description of diffusion-reaction processes and their first-passage statistics, with the special emphasis on the role of the boundary local time and related spectral tools. The second part of the talk will illustrate the use of these tools for the analysis of boundary-catalytic branching processes. These processes describe a broad class of natural phenomena where the population of diffusing particles grows due to their spontaneous binary branching (e.g., division, fission, or splitting) on a catalytic boundary located in a complex environment. We investigate the possibility of the geometric control of the population growth by compensating the proliferation of particles due to catalytic branching events by their absorptions in the bulk or on boundary absorbing regions. We identify an appropriate Steklov spectral problem to obtain the phase diagram of this out-of-equilibrium stochastic process. The principal eigenvalue determines the critical line that separates an exponential growth of the population from its extinction. In other words, we establish a powerful tool for calculating the optimal absorption rate that equilibrates the opposite effects of branching and absorption events and thus results in steady-state behavior of this diffusion-reaction system. Moreover, we show the existence of a critical catalytic rate above which no compensation is possible, so that the population cannot be controlled and keeps growing exponentially. The proposed framework opens promising perspectives for better understanding, modeling, and control of various boundary-catalytic branching processes, with applications in physics, chemistry, and life sciences.

Vertex operator algebras provide a succinct mathematical description of the chiral sector of two-dimensional conformal field theories. Various extensions of the framework of vertex operator algebras have been proposed in the literature which are capable of describing full two-dimensional conformal field theories, including both chiral and anti-chiral sectors. I will explain how the notion of a full vertex operator algebra can be elegantly described using the modern language of factorisation algebras developed by Costello and Gwilliam. This talk will be mainly based on [arXiv:2501.08412].

In this week's Fridays@2, a panel of representatives from a range of companies who employ mathematics graduates will be here to answer your questions.A degree in mathematics opens doors far beyond academia, but what do those paths really look like? Join us for a panel event bringing together mathematicians working across Finance, Digital Services, Technology, Consulting, Data Analytics, and Teaching.

Our speakers will share their career journeys, how they moved from studying mathematics into industry roles, and what their day to day work involves. This is your opportunity to gain insight into the skills employers value, the challenges and opportunities in different sectors, and the many ways mathematical thinking shapes real world impact.

Whether you already have a clear goal or are still exploring your options, come along with your questions and curiosity and discover where maths could take you.

I will present joint work with M. Botnan, S. Oppermann, and J. Steen on multiparameter persistent homology in degree zero. It is known that arbitrary diagrams of vector spaces and linear maps can be realized as homology of diagrams of simplicial complexes in some positive degree. We study the more restrictive case of degree zero, which corresponds to diagrams freely generated from sets and set maps. Despite their seemingly simple combinatorial nature, a full understanding of the structure of these representations remains elusive. I will summarize our findings and discuss some conjectures.

I’ll discuss a top down example of holography at zero ’t Hooft coupling. The gauge side is self-dual N=4 SYM, whilst the gravitational side is the closed topological B-model on a certain Calabi-Yau 7-fold that fibres over twistor space. As an application, I’ll discuss an interpretation of correlation functions of certain determinant operators in sI’ll discuss a top down example of holography at zero ’t Hooft coupling. The gauge side is self-dual N=4 SYM, whilst the gravitational side is the closed topological B-model on a certain Calabi-Yau 7-fold that fibres over twistor space. As an application, I’ll discuss an interpretation of correlation functions of certain determinant operators in self-dual SYM in terms of giant graviton D-branes on this 7-fold.elf-dual SYM in terms of giant graviton D-branes on this 7-fold.

Joint with Atul Sharma arxiv:2512.04152.

We survey the life of a Turing pattern, from initial diffusive instability through the emergence of dominant spatial modes and to an eventual spatially heterogeneous pattern. While many mathematically ideal Turing patterns are regular, repeating in structure and remaining of a fixed length scale throughout space, in the real world there is often a degree of irregularity to patterns. Viewing the life of a Turing pattern through the lens of spatial modes generated by the geometry of the bounded space domain housing the Turing system, we discuss how irregularity in a Turing pattern may arise over time due to specific features of this space domain or specific spatial dependencies of the reaction-diffusion system generating the pattern.

Shimura varieties are highly symmetric algebraic varieties that play an important role in the Langlands program. In the first part of the talk, I will try to give you a sense of what they are like, with a focus on their different kinds of symmetries. In the second part of the talk, I will introduce Igusa stacks, a powerful new tool in the study of Shimura varieties. To illustrate their role, I will discuss how Igusa stacks can shed light on the many structures that exist on the intersection cohomology of Shimura varieties. This is joint work in progress with Linus Hamann and Mingjia Zhang.

In the past years, deep learning algorithms have been applied to numerous classical problems from mathematical finance. In particular, deep learning has been employed to numerically solve high-dimensional derivatives pricing and hedging tasks. Theoretical foundations of deep learning for these tasks, however, are far less developed. In this talk, we start by revisiting deep hedging and introduce a recently developed adversarial training approach for making it more robust. We then present our recent results on theoretical foundations for approximating option prices, solutions to jump-diffusion PDEs and optimal stopping problems using (random) neural networks, allowing to obtain more explicit convergence guarantees. We address neural network expressivity, highlight challenges in analysing optimization errors and show the potential of random neural networks for mitigating these difficulties.

Dr Carolina Urzua Torres will talk about 'Paving the way to a T-coercive method for the wave equation'

Space-time Galerkin methods are gradually becoming popular, since they allow adaptivity and parallelization in space and time simultaneously. A lot of progress has been made for parabolic problems, and its success has motivated an increased interest in finding space-time formulations for the wave equation that lead to unconditionally stable discretizations. In this talk I will discuss some of the challenges that arise and some recent work in this direction.

In particular, I will present what we see as a first step toward introducing a space-time transformation operator $T$ that establishes $T$-coercivity for the weak variational formulation of the wave equation in space and time on bounded Lipschitz domains. As a model problem, we study the ordinary differential equation (ODE) $u'' + \mu u = f$ for $\mu>0$, which is linked to the wave equation via a Fourier expansion in space. For its weak formulation, we introduce a transformation operator $T_\mu$ that establishes $T_\mu$-coercivity of the bilinear form yielding an unconditionally stable Galerkin-Bubnov formulation with error estimates independent of $\mu$. The novelty of the current approach is the explicit dependence of the transformation on $\mu$ which, when extended to the framework of partial differential equations, yields an operator acting in both time and space. We pay particular attention to keeping the trial space as a standard Sobolev space, simplifying the error analysis, while only the test space is modified.
The theoretical results are complemented by numerical examples.  

The computation of accurate low-rank matrix approximations is central to improving the scalability of various techniques in machine learning, uncertainty quantification, and control. Traditionally, low-rank approximations are constructed using SVD-based approaches such as truncated SVD or RandomizedSVD. Although these SVD approaches---especially RandomizedSVD---have proven to be very computationally efficient, other low-rank approximation methods can offer even greater performance. One such approach is the CUR decomposition, which forms a low-rank approximation using direct row and column subsets of a matrix. Because CUR uses direct matrix subsets, it is also often better able to preserve native matrix structures like sparsity or non-negativity than SVD-based approaches and can facilitate data interpretation in many contexts. This paper introduces IterativeCUR, which draws on previous work in randomized numerical linear algebra to build a new algorithm that is highly competitive compared to prior work: (1) It is adaptive in the sense that it takes as an input parameter the desired tolerance, rather than an a priori guess of the numerical rank. (2) It typically runs significantly faster than both existing CUR algorithms and techniques such as RandomizedSVD, in particular when these methods are run in an adaptive rank mode. Its asymptotic complexity is  $\mathcal{O}(mn + (m+n)r^2 + r^3)$ for an $m\times n$ matrix of numerical rank $r$. (3) It relies on a single small sketch from the matrix that is successively downdated as the algorithm proceeds.

You will likely be familiar with the notion of a hydrogen atom, having seen something about its discrete energy levels and orbitals at some point or another. This is an example of a quantum system. In this talk, we explore what transpires when taking a quantum system and placing it into a three-dimensional container having some prescribed geometry. In the limit where the container is large (relative to the natural lengthscale of the quantum system), its influence over the quantum system is negligible; yet, as the container is made small (comparable to the aforementioned lengthscale), geometric information intrinsic to the container plays an important role in determining the energy and orbital structure of the system. We describe how to do (numerically-assisted) perturbation theory in this small-container limit and then match it to the large-box regime, using a combination of these asymptotics and direct simulations to tell the story of geometrically confined quantum systems. Much of our focus will be on linear Schrödinger equations governing single-particle quantum systems; however, time permitting, we will briefly discuss how to do similar things to study geometrically confined nonlinear Schrödinger equations, with geometric confinement of Bose-Einstein condensates being a primary motivation. Geometric confinement of an attractive Bose-Einstein condensate can, for instance, modify the collapse threshold and enhance stability, with the particular choice of confining geometry shifting the boundary of instability, staving off the collapse which is prevalent in three-dimensional attractive condensates.

A first-order theory $T$ is a model-complete core theory if every first-order formula is equivalent modulo $T$ to an existential positive formula; a core companion of a theory $T$ is a model-complete core theory $S$ such that every model of $T$ maps homomorphically to a model of $S$ and vice-versa. Whilst core companions may not exist in general, if they exist, they are unique. Moreover, $\omega$-categorical theories always have a core companion, which is also $\omega$-categorical.

In the first part of this talk, we show that many model-theoretic properties, such as stability, NIP, simplicity, and NSOP, are preserved when moving to the core companion of a complete theory.

In the second part of this talk, we study the notion of core interpretability, which arises by taking the core companions of structures interpretable in a given structure. We show that there are structures which are core interpretable but not interpretable in $(\mathbb{N};=)$ or $(\mathbb{Q};<)$. We conjecture that the class of structures which are core interpretable in $(\mathbb{N};=)$ equals the class of $\omega$-stable first-order reducts of finitely homogeneous relational structures, which was studied by Lachlan in the 80's. We present some partial results in this direction, including the answer a question of Walsberg.

This is joint work with Manuel Bodirsky and Bertalan Bodor.

 Given a group acting on a metric space X, one is often interested in the kernel of the action, consisting of those elements that fix every point of X. From a coarse geometric perspective, however, this notion is unsatisfactory, as the kernel is generally not invariant under G-equivariant quasi-isometries. To address this, one can instead consider the coarse kernel, defined as the collection of group elements that move every point of X by a uniformly bounded amount. In this talk, we study this coarse kernel under various assumptions on the action. 

When the action is geometric, we give a purely algebraic characterisation of the coarse kernel as the FC-centre of the group. We then specialise to actions on CAT(0) spaces, where we investigate the coarse kernel via the curtain model, a hyperbolic space associated to a CAT(0) space introduced by Petyt, Spriano, and Zalloum. Along the way, we will meet centralisers, boundaries, and actions on hyperbolic spaces! This is based on my summer project supervised by Davide Spriano and Harry Petyt.