Past

How do you efficiently look for jobs?

How can you make the most of careers fairs?

What makes a CV or cover letter stand out?

Get practical advice and bring your questions!

In this talk I will give a gentle introduction to some aspects of the theory of hypergeometric functions as a natural language for addressing various integrals appearing in quantum field theory (QFT). In particular I will focus on the so-called intersection pairings as well as the differential equations satisfied by the integrals, and I will show how these aspects of the mathematical theory can find a natural interpretation in concrete QFT applications. I will mostly focus on Feynman integrals as paradigmatic example, where the language will shed new light on our most powerful method for computing Feynman integrals as well as their non-local symmetries. I will then give an outlook how these methods could allow us to also learn about integrals appearing in other places in field and string theory, such as Coulomb branch amplitudes, celestial holography and AdS (supergravity and string) amplitudes.

There will be no journal club this week to avoid conflicting with FPUK.

Over the last decade, adversarial attack algorithms have revealed instabilities in artificial intelligence (AI) tools. These algorithms raise issues regarding safety, reliability and interpretability; especially in high risk settings. Mathematics is at the heart of this landscape, with ideas from  numerical analysis, optimization, and high dimensional stochastic analysis playing key roles. From a practical perspective, there has been a war of escalation between those developing attack and defence strategies. At a more theoretical level, researchers have also studied bigger picture questions concerning the existence and computability of successful attacks. I will present examples of attack algorithms for neural networks in image classification, for transformer models in optical character recognition and for large language models. I will also show how recent generative diffusion models can be used adversarially. From a more theoretical perspective, I will outline recent results on the overarching question of whether, under reasonable assumptions, it is inevitable that AI tools will be vulnerable to attack.

This talk will focus on various definitions of orientability for non-smooth spaces with Ricci curvature bounded from below. The stability of orientability and non-orientability will be discussed. As an application, we will prove the orientability of 4-manifolds with non-negative Ricci curvature and Euclidean volume growth. This work is based on a collaboration with E. Bruè and A. Pigati.

We apply techniques of exponential asymptotics to the KdV equation to derive the small-time behaviour for dispersive waves that propagate in one direction.  The results demonstrate how the amplitude, wavelength and speed of these waves depend on the strength and location of complex-plane singularities of the initial condition.  Using matched asymptotic expansions, we show how the small-time dynamics of complex singularities of the time-dependent solution are dictated by a Painlevé II problem with decreasing tritronquée solutions.  We relate these dynamics to the solution on the real line.

This talk focuses on the logic side of the following result: the non-definability of free independence in the theory of tracial von Neumann algebras and C*-probability spaces. I will introduce continuous model theory, which is suitable for the study of metric structures. Definability in the continuous setting differs slightly from that in the discrete case. I will introduce its definition, give examples of definable sets, and prove an equivalent ultrapower condition of it. A. Berenstein and C. W. Henson exposited model theory for probability spaces in 2023, which was done with continuous model theory. It makes it natural for us to consider the definability of the notion of free independence in probability spaces. I will explain our result, which gives an example of a non-definable set.

This is work with William Boulanger and Emma Harvey, supervised by Jenny Pi and Jakub Curda.

Come take a break and get to know other Mathematrix members over some crafts! All supplies and sweet treats provided.

We derive a lower bound, independent of the initial condition, for the solution of the KPZ equation on the torus, using its representation as the value function of a stochastic control problem.

With the same techniques we also prove a bound for its oscillation, again independent of initial conditions, which is related to Harnack's inequality for the (rough) heat equation.

I will review how the equations of general relativity near a spacetime singularity map onto an arithmetic hyperbolic billiard dynamics. The semiclassical quantum states for this dynamics are Maaβ cusp forms on fundamental domains of modular groups. For example, gravity in four spacetime dimensions leads to PSL(2,Z) while five dimensional gravity leads to PSL(2,Z[w]), with Z[w] the Eisenstein integers. The automorphic forms can be expressed, in a dilatation (Mellin transformed) basis as L-functions. The Euler product representation of these L-functions indicates that these quantum states admit a dual interpretation as a "primon gas" partition function. I will describe some physically motivated mathematical questions that arise from these observations.

I will review how the equations of general relativity near a spacetime singularity map onto an arithmetic hyperbolic billiard dynamics. The semiclassical quantum states for this dynamics are Maaβ cusp forms on fundamental domains of modular groups. For example, gravity in four spacetime dimensions leads to PSL(2,Z) while five dimensional gravity leads to PSL(2,Z[w]), with Z[w] the Eisenstein integers. The automorphic forms can be expressed, in a dilatation (Mellin transformed) basis as L-functions. The Euler product representation of these L-functions indicates that these quantum states admit a dual interpretation as a "primon gas" partition function. I will describe some physically motivated mathematical questions that arise from these observations.

In this talk, Aaron Kettner, Institute of Mathematics, Czech Academy of Sciences, will show how to construct a C*-correspondence from a vector bundle together with a (partial) homeomorphism on the bundle's base space. The associated Cuntz-Pimsner algebras provide a class of examples that is both tractable and potentially quite large. Under reasonable assumptions, these algebras are classifiable in the sense of the Elliott program. If time permits, Aaron will sketch some K-theory calculations, which are work in progress.

This illustrated historical talk covers the period from around 1890, when graph theory was still mainly a collection of isolated results, to the 1990s, when it had become part of mainstream mathematics. Among many other topics, it includes material on graph and map colouring, factorisation, trees, graph structure, and graph algorithms. 

In this talk, I will introduce a new framework for working with moduli stacks in enumerative geometry, aimed at generalizing existing theories of enumerative invariants counting objects in linear categories, such as Donaldson–Thomas theory, to general, non-linear moduli stacks. This involves a combinatorial object called the component lattice, which is a globalization of the cocharacter lattice and the Weyl group of an algebraic group.

Several important results and constructions known in linear enumerative geometry can be extended to general stacks using this framework. For example, Donaldson–Thomas invariants can be defined for a general class of stacks, not only linear ones such as moduli stacks of sheaves. As another application, under certain assumptions, the cohomology of a stack, which is often infinite-dimensional, decomposes into finite-dimensional pieces carrying enumerative information, called BPS cohomology, generalizing a result of Davison–Meinhardt in the linear case.

This talk is based on joint works with Ben Davison, Daniel Halpern-Leistner, Andrés Ibáñez Núñez, Tasuki Kinjo, and Tudor Pădurariu.

Even in a totally discrete space $X$ you need to know how to move between distinct points. A path $P_{x,y}$ between two points $x,y \in X$ is a sequence of points in $X$ that starts with $x$ and ends with $y$. A path system is a collection of paths $P_{x,y}$, one per each pair of distinct points $x, y$ in $X$. We restrict ourselves to the undirected case where $P_{y,x}$ is $P_{x,y}$ in reverse.

Strictly metrical path systems are ubiquitous. They are defined as follows: There is some spanning, connected graph $(X, E)$ with positive edge weights $w(e)$ for all $e\in E$ and $P_{x,y}$ is the unique $w$-shortest $xy$ path. A metrical path system is defined likewise, but $w$-shortest paths need not be unique. Even more generally, a path system is called consistent  (no $w$ is involved here) if it satisfies the condition that when point $z$ is in $P_{x,y}$, then $P_{x,y}$ is $P_{x,z}$ concatenated with $P_{z,y}$. These three categories of path systems are quite different from each other and in our work we find quantitative ways to capture these differences.

Joint work with Daniel Cizma.

Recent foundational work by Ben-Bassat, Kelly, and Kremnitzer describes a model for derived analytic geometry as homotopical geometry relative to the infinity category of simplicial commutative complete bornological rings. In this talk, Rhiannon Savage will discuss a representability theorem for derived stacks in these contexts and will set out some new foundations for derived smooth geometry. Rhiannon will also briefly discuss the representability of the derived moduli stack of non-linear elliptic partial differential equations by an object we call a derived C∞-bornological affine scheme.

This tutorial will be a hands-on introduction to proving theorems in Lean, using its mathematical library Mathlib. It will not assume any previous knowledge about formal theorem provers. We will discover the Lean language, learn how to read a statement and a proof, and learn the essential "tactics" one can use to prove theorems in Lean.

Participants should come with a computer, and it would be best if they could install Lean before the tutorial by following the instructions at https://lean-lang.org/install/ . The installation should be easy and takes only a few minutes.

I will explain how the irreversibility of the renormalization group together with anomalies, including anomalies in the space of coupling constants, can be used to constrain the IR phases of defects in familiar quantum field theories. As an example, I will use these techniques to provide evidence for a conjectural "spin-flux duality" which describes how certain line operators are mapped across particle/vortex duality in 2+1d.

Einstein’s theory of general relativity reshaped our understanding of the universe. Instead of thinking of gravity as a force, Einstein showed it is the bending and warping of space and time caused by mass and energy. This radical idea not only explained how planets orbit stars, but also opened the door to astonishing predictions. In this seminar we will explore some of its most fascinating consequences from the expansion of the universe, to gravitational waves, and the existence of black holes.

The Ginzburg–Landau energy is often used to approximate the Dirichlet energy. As the perturbation parameter tends to zero, critical points of the Ginzburg–Landau energy converge, in an appropriate (bubbling) sense, to harmonic maps. In this talk I will first explain key analytical properties of this approximation procedure, then show that not every harmonic map can be approximated in this way. This is based on a rigidity theorem: under the energy threshold of 8pi, we classify all solutions of the associated nonlinear elliptic system from S2 to S2, thereby identifying exactly which harmonic maps can arise as Ginzburg–Landau limits in this regime.

A conjecture by Malle gives a prediction for the number of number fields of bounded discriminant. In this talk I will give an asymptotic formula for the number of abelian number fields of bounded height whose ramification type has been restricted to lie in a given subset of the Galois group and provide an explicit formula for the leading constant. I will then describe how counting these number fields can be viewed as a problem of counting rational points on the stack BG and how the existence of such number fields is controlled by a Brauer-Manin obstruction. No prior knowledge of stacks is needed for this talk!

TBC

I will present a collaborative project in which we formalized the construction of Brownian motion in Lean. Lean is an interactive theorem prover, with a large mathematical library called Mathlib. I will give an introduction to Lean and Mathlib, explain why one may want to formalize mathematics, and give a tour of the probability theory part of Mathlib. I will then describe the Brownian motion project, its organization, and some of the formalized results. For that project, we developed the theory of Gaussian measures and implemented a proof of Kolmogorov's extension theorem, as well as a modern version of the Kolmogorov-Chentsov continuity theorem based on Talagrand's chaining technique. Finally, I will discuss the next step of the project: formalizing stochastic integrals.

Symplectic implosion is an abelianisation construction in symplectic geometry. The implosion of the cotangent bundle of a group K plays a universal role in the implosion of manifolds with a K-action.  This universal implosion, which is usually a singular variety, can also be viewed as the non-reductive Geometric Invariant Theory quotient of the complexification G of K by its maximal unipotent subgroup. 

In this talk, we describe joint work with Johan Martens and Nick Proudfoot which uses point-counting techniques to calculate the intersection cohomology of the universal implosion.

Quantum tomography involves obtaining a full classical description of a prepared quantum state from experimental results. We propose a Langevin sampler for quantum tomography, that relies on a new formulation of Bayesian quantum tomography exploiting the Burer-Monteiro factorization of Hermitian positive-semidefinite matrices. If the rank of the target density matrix is known, this formulation allows us to define a posterior distribution that is only supported on matrices whose rank is upper-bounded by the rank of the target density matrix. Conversely, if the target rank is unknown, any upper bound on the rank can be used by our algorithm, and the rank of the resulting posterior mean estimator is further reduced by the use of a low-rank promoting prior density. This prior density is a complex extension of the one proposed in [Annales de l’Institut Henri Poincaré Probability and Statistics, 56(2):1465–1483, 2020]. We derive a PAC-Bayesian bound on our proposed estimator that matches the best bounds available in the literature, and we show numerically that it leads to strong scalability improvements compared to existing techniques when the rank of the density matrix is known to be small.

I will review notions of categorical complexity, and the more recent work of Biran, Cornea and Zhang on fragmentation in triangulated persistence categories (TPCs), then go on to discuss applications of this to filtered topology. In particular, we will consider a suitable category of filtered topological spaces and detail some constructions and properties, before showing that an associated 'filtered stable homotopy category' is a TPC. I will then give some interesting results relating to this.

Prof Ian Griffiths (a mental health first aider in the department) will lead a discussion about how to protect your mental health when studying an intense graduate degree and outline the support and resources available within the Mathematical Institute. 

We'll discuss what mathematicians are looking for in written solutions.  How can you set out your ideas clearly, and what are the standard mathematical conventions?

This session is likely to be most relevant for first-year undergraduates, but all are welcome.

Motivated by pattern formations and cell movements, many evolution equations incorporating spatial convolution with suitable integral kernel have been proposed. Numerical simulations of these nonlocal evolution equations can reproduce various patterns depending on the shape and form of integral kernel.The solutions to nonlocal evolution equations are similar to the patterns obtained by reaction-diffusion system and Keller-Segel system models. In this talk, we classify nonlocal interactions into two types, and investigate their relationship with reaction-diffusion systems and Keller-Segel systems, respectively. In these partial differential equation systems, we introduce multiple auxiliary diffusive substances and consider the singular limit of the quasi-steady state to approximate nonlocal interactions. In particular, we introduce how the parameters of the partial differential equation system are determined by the given integral kernel. These analyses demonstrate that, under certain conditions, nonlocal interactions and partial differential equation systems can be treated within a unified framework.  
This talk is based on collaborations with Hiroshi Ishii of Hokkaido University and Hideki Murakawa of Ryukoku University. 

We develop a general framework for pathwise stochastic integration that extends Follmer's classical approach beyond gradient-type integrands and standard left-point Riemann sums and provides pathwise counterparts of Ito, Stratonovich, and backward Ito integration. More precisely, for a continuous path admitting both quadratic variation and Levy area along a fixed sequence of partitions, we define pathwise stochastic integrals as limits of general Riemann sums and prove that they coincide with integrals defined with respect to suitable rough paths. Furthermore, we identify necessary and sufficient conditions under which the quadratic variation and the Levy area of a continuous path are invariant with respect to the choice of partition sequences.

Given a von Neumann algebra M, we can consider the set of values of p such that Lp(M) has the approximation property: the identity on it is a limit of finite rank operators for a suitable topology. Apart from the case when p is infinite, which has been the subject of a lot of work initiated by Haagerup in the late 70s, this invariant has not been very much exploited so far. But ancient works in collaboration with Vincent Lafforgue and Tim de Laat suggest that, maybe, it can distinguish the factors of SL(n,Z) for different values of n. I will explain something that I realized only recently, and that explains why this is a difficult question: it implies some form of the classical Kakeya conjecture, which predicts the shape of sets in the Euclidean space in which a needle can be turned upside down. This talk from Mikael de la Salle will be an opportunity to discuss other connections between classical Fourier analysis and analysis in group von Neumann algebras, including in collaboration with Javier Parcet and Eduardo Tablate

We'll describe connections between probabilistic models for primes,
the Hardy-Littlewood k-tuples conjectures, the distribution of primes in
very short intervals, the interval sieve, and hypothetical Landau-Siegel
zeros of Dirichlet L-functions.  We will emphasize the role and limitations
of probabilistic ideas.

I will discuss a two-dimensional dilaton gravity theory with a sine potential. At the disk level, this theory admits a microscopic holographic realization as the double-scaled SYK model. Remarkably, in the open channel canonical quantization of the theory, the momentum conjugate to the length of two-sided Cauchy slices becomes periodic. As a result, the ERB length in sine dilaton gravity is discretized upon gauging this symmetry. For closed Cauchy slices, a similar discretization occurs in the physical Hilbert space, corresponding to a discrete spectrum for the length of the necks of trumpet geometries. By appropriately gluing two such trumpets together, one can then construct a wormhole geometry in sine dilaton gravity, whose amplitude matches the spectral correlation functions of a one-cut matrix integral. This correspondence suggests that the theory provides a path integral formulation of q-deformed JT gravity, where the matrix size is large but finite. Finally, I will describe how this theory of gravity can be regarded as a realization of q-deformed holography and propose a possible implementation of this framework to study the near-horizon dynamics of near-extremal de Sitter black holes.

Time-series datasets are central in numerous fields of science and engineering, such as biomedicine, Earth observation, and network analysis. Extensive research exists on state-space models (SSMs), which are powerful mathematical tools that allow for probabilistic and interpretable learning on time series. Estimating the model parameters in SSMs is arguably one of the most complicated tasks, and the inclusion of prior knowledge is known to both ease the interpretation but also to complicate the inferential tasks. In this talk, I will introduce a novel joint graphical modeling framework called DGLASSO (Dynamic Graphical Lasso) [1], that bridges the static graphical Lasso model [2] and the causal-based graphical approach for the linear-Gaussian SSM in [3]. I will also present a new inference method within the DGLASSO framework that implements an efficient block alternating majorization-minimization algorithm. The algorithm's convergence is established by departing from modern tools from nonlinear analysis. Experimental validation on synthetic and real weather variability data showcases the effectiveness of the proposed model and inference algorithm.

[1] E. Chouzenoux and V. Elvira. Sparse Graphical Linear Dynamical Systems. Journal of Machine Learning Research, vol. 25, no. 223, pp. 1-53, 2024

[2] J. Friedman, T. Hastie, and R. Tibshirani. Sparse inverse covariance estimation with the graphical LASSO. Biostatistics, vol. 9, no. 3, pp. 432–441, Jul. 2008.

[3] V. Elvira and E. Chouzenoux. Graphical Inference in Linear-Gaussian State-Space Models. IEEE Transactions on Signal Processing, vol. 70, pp. 4757-4771, Sep. 2022.

In the numerical simulation of incompressible flows and elastic materials, it is often desirable to design discretisation schemes that preserve key structural properties of the underlying physical model. In particular, the conservation of angular momentum plays a critical role in accurately capturing rotational effects, and is closely tied to the symmetry of the stress tensor. Classical formulations such as the Stokes equations or linear elasticity can exhibit significant discrepancies when this symmetry is weakly enforced or violated at the discrete level.

This work focuses on mixed finite element methods that impose the symmetry of the stress tensor strongly, thereby ensuring exact conservation of angular momentum in the absence of body torques and couple stresses. We systematically study the effect of this constraint in both incompressible Stokes flow and linear elasticity, including anisotropic settings inspired by liquid crystal polymer networks. Through a series of benchmark problems—ranging from rigid body motions to transversely isotropic materials—we demonstrate the advantages of angular-momentum-preserving discretisations, and contrast their performance with classical elements.

Our findings reveal that strong symmetry enforcement not only leads to more robust a priori error estimates and pressure-independent velocity approximations, but also more reliable physical predictions in scenarios where angular momentum conservation is critical.

These insights advocate for the broader adoption of structure-preserving methods in computational continuum mechanics, especially in applications sensitive to rotational invariants.

When a specimen of non-trivial shape undergoes deformation under a dead load or during an active process, finite element simulations are the only technique for evaluating the deformation. Classical books describe complicated techniques for evaluating stresses and strains in semi-infinite, circular or cylindrical objects.  However, the results obtained are limited, and it is well known that elasticity (linear or nonlinear) is strongly intertwined with geometry. For the simplest geometries, it is possible to determine the exact deformation, essentially for low loading values, and prove that there is a threshold above which the specimen loses stability. The next step is to apply perturbation techniques (linear and nonlinear bifurcation theory).
 

In this talk, I will demonstrate how many aspects can be simplified or revealed through the use of complex analysis and conformal mapping techniques for shapes, strains, and active stresses in thin samples. Examples include leaves and embryonic jellyfish.

I will talk about some recent work with Tingxiang Zou on higher-dimensional Elekes-Szabó problems in the case of an Ind-constructible action of a group G on a variety X. We expect nilpotent algebraic subgroups N of G to be responsible for any such; this roughly means that if H and A are finite subsets with non-expansion |H*A| <= |A|^{1+\eta}, then H concentrates on a coset of some such N.

A natural example is the action of the Cremona group of birational transformations of the plane. I will talk about a recent result which confirms the above expectation when we restrict to the group of polynomial automorphisms of the plane, using Jung's description of this group as an amalgamated free product, as well as some work in progress which combines weak polynomial Freiman-Ruzsa with effective Mordell-Lang, after Akshat Mudgal, to handle some further special cases.

A 2024 collection of articles in the Bulletin of the AMS asked "Will machines change mathematics?", suggesting that  "Pure mathematicians are used to enjoying a great degree of research autonomy and intellectual freedom, a fragile and precious heritage that might be swept aside by a mindless use of machines." and challenging readers to  "decide upon our subject’s future direction.”

This was a response to the mathematical capabilities of emerging technologies, alone or in combination. These techniques include  software such as LEAN for  providing formal proofs; use of LLMs to produce credible, if derivative, research papers with expert human guidance; specialist algorithms such as AlphaGeometry; and sophisticated use of machine learning to search for examples.  Their development (at huge cost in compute power and energy) has been accompanied by an unfamiliar and exuberant level of hype from well-funded start-ups claiming to “solve mathematics” and the like. And it raises questions beyond the technical concerning governance, funding and the nature of the mathematical profession.

To try and understand what’s going on we look historical examples of changes in mathematical practice - as an example we consider key developments in the early days of computational group theory.

The speaker is keen to hear of colleagues using LLMs, LEAN or similar things in research, even if they can’t come to the talk.