Skip to content
Technical University of Munich

Workshop: Probability and Statistical Physics

Probability Theory |

The workshop aims at bringing together speakers in Probability and Statistical Physics, in order to discuss advances in the field. In this sense, we plan to have 13 talks by renowned speakers. The workshop starts on Wednesday afternoon and ends on Friday at lunchtime.

The scientific program is organized by Diana Conache, Dirk-André Deckert, Silke Rolles.

There is no conference fee and everybody is welcome to attend. In particular, all the members of SPP 2265 are invited.

The workshop is partially funded by the DFG priority program SPP 2265 Random Geometric System. Members of the SPP who are interested in participating in the workshop should send an email to sekretariat(at)wt.cit.tum.de. 

Location and dates

The workshop takes place from October 7th (Wednesday) to October 9th (Friday) 2026, at the Mathematics Institute of the LMU Munich. All talks take place in room B 005. The coffee breaks take place in room B 349.

How to get there from Munich Central Station

From Munich Central Station, take bus 100 or 58 from “Hauptbahnhof Nord” and get off at “Pinakotheken.” The LMU Mathematics Institute is located close to the bus stop at Theresienstraße 39. Alternatively, the institute can be reached on foot from Munich Central Station in approximately 20 minutes. For current connections and service information, please consult the MVG journey planner.

Location of the lecture and coffee break rooms

The lecture room is located on the ground floor of the institute, next to one of the entrances on Theresienstraße. Click here for the precise location inside the building. The coffee break room is located on the third floor and is easily accessible by elevator or stairs. Click here for the precise locaton inside the building.

Program

Wednesday, October 7, 2026

  • 14:00-14:10 Opening
  • 14:10-15:00 Frank den Hollander
  • 15:00-15:50 Xiaolin Zeng
  • 15:50-16:20 Coffee Break
  • 16:20-17:10 Torben Krüger
  • 17:10-18:00 Mario Wüthrich 

Thursday, October 8, 2026

  • 09:00-09:50 Roland Bauerschmidt
  • 09:50-10:40 Pierre Tarrès
  • 10:40-11:10 Coffee Break
  • 11:10-12:00 Margherita Disertori
  • 12:00-14:00 Lunch
  • 14:00-14:50 Dirk-André Deckert
  • 14:50-15:20 Coffee Break
  • 15:20-18:00 Get together in honor of Prof. Franz Merkl
  • 18:00 Reception for workshop participants

Friday, October 9, 2026

  • 09:00-09:50 Christophe Sabot
  • 09:50-10:40 Thomas Richthammer
  • 10:40-11:10 Coffee Break
  • 11:10-12:00 Martin Zerner
  • 12:00-12:50 Friedrich Götze
  • 12:50-14:00 Lunch

Speakers

 

Abstracts

Roland Bauerschmidt -- The sinh-Gordon theory

I will discuss the probabilistic construction and features of the sinh-Gordon quantum field theory.

Dirk-André Deckert -- The Secret Life of Franz, the Physicist

Let me take you on a journey through the last two decades and show you a perhaps lesser-known side of Franz Merkl's work: his research on the foundations of quantum electrodynamics. I will keep this talk accessible without requiring a deeper physics background and illustrate some of the mathematical challenges and beautiful structures encountered.

Margherita Disertori -- Fractional Anderson model and long-range self-avoiding walks

The fractional Anderson model has been subject to increasing interest in recent years in the context of materials with anomalous diffusion. In this context we adapted a strategy, proposed by Schenker in 2015, relating the averaged Green’s function to the two point function of a self-avoiding walk with long jumps, the key tool being a depleted resolvent identity.

This is joint work with Constanza Rojas-Molina and Roberto Maturana

Friedrich Götze -- Measure Transport Distances and Energy Bounds

We discuss recent results on tight bounds for Kantorovich and Zolotarev distances of measures in Euclidean spaces in terms of their one dimensional projections. Furthermore we introduce transport distances in Euclidean spaces based on classes of convex loss functions which extend the standard Wasserstein metrics. For this class we derive energy bounds in terms of  their dual Sobolev norms.  This is joint work with Sergey Bobkov.

Frank den Hollander -- Self-reinforced preferential attachment

We consider a preferential attachment random graph with self-reinforcement. Each time a new vertex comes in, it attaches itself to an old vertex with a probability that is proportional to the sum of the degrees of that old vertex at all prior times. The resulting growing graph is a random tree whose vertices have degrees that grow polynomially fast in time. We compute the growth exponent, show that it is strictly larger than the growth exponent in the absence of self-reinforcement, and develop insight into how the self-reinforcement affects the growth. Proofs are based on a stochastic approximation scheme.

Torben Krüger – Spectral singularities of structured random matrices

As the dimension of a random matrix tends to infinity, its empirical eigenvalue distribution often converges to a deterministic limiting measure. For Hermitian random matrices this measure is supported on the real line, while for non- Hermitian matrices it is supported on the complex plane. In the basic Hermitian case of a matrix with independent, identically distributed entries above the diagonal, the limiting spectral measure is the famous semicircle law. Its density exhibits two standard types of local behavior: it is positive in the bulk of the spectrum, while at the spectral edges it vanishes with square-root behavior. Correspondingly, on the local eigenvalue-spacing scale, the eigenvalues are governed by two universal limit distributions: the sine kernel statistics in the bulk and the Airy kernel at the edge. When the distribution of the matrix entries has additional structure, for instance when the entries have non-identical variances, the limiting spectral density may develop further singularities beyond the usual bulk and edge regimes. These singularities lead to new local scaling limits and new universal eigenvalue statistics. In this talk, we give an overview of recent progress on the classification of such spectral singularities and on the corresponding local universal laws.

Thomas Richthammer -- Conservation and breaking of symmetries in the hard disk model and similar 2D models

We consider random distributions of particles in 2D with hard constraints. Our main interest is in the hard disk model, which can be thought of a Poisson point process in the plane conditioned on every pair of particles having a distance of at least 1. This can be made rigorous in terms of a Gibbsian point process. We will give an overview over some known results and open problems on this process and in particular present results on the conservation of translational symmetry. We will also give an overview over other interesting examples of such processes with hard constraints and in particular look at the Boolean model conditioned on having one connected component (which in some sense is a Poisson point process with the opposite conditioning as that in the hard disk model).

Christophe Sabot – A random polymer approach to the weak disorder phase of the Vertex Reinforced Jump Process

The Vertex Reinforced Jump Process (VRJP) is a continuous-time process closely related to the linearly edge reinforced random walk. The recurrence/transience of the VRJP can be caracterized by the asymptotic behavior of a positive martingale : the VRJP is recurrent when the limit is null and transient when it is positive. Besides, the L^p integrability of that martingale is related to the diffusive behavior of the VRJP.
We will recall how that martingale appears and explain how it can be interpreted as the partition function of a non-directed polymer in a very specific 1-dependent potential. We will show some results about the L^p integrability of the martingale, using the polymer interpretation and taking inspiration from some works of Junk and Lacoin on directed polymers. 

Based on a joint work with Q. Berger, A. Legrand and R. Poudevigne-Auboiron.

Pierre Tarrès -- Once-reinforced random walk

We discuss new developments on self-interacting random walks, such as recurrence/transience, range at a given time, and local time formulas, in particular for the once-reinforced random walk (ORRW). 

Mario Wüthrich -- Actuarial Modeling in the Era of AI

Starting from classical actuarial pricing based on statistical modeling, we trace the evolution toward modern AI methods. We first show how generalized linear models can be extended to neural networks, providing a natural bridge from traditional actuarial modeling to representation learning. We then move toward large language model architectures by introducing contextualized embeddings based on transformers. Next, we extend these models to in-context learning, allowing them to learn from similar instances without explicit model retraining. Finally, we introduce a reasoning layer that successively refines the learned embeddings. This progression illustrates how established actuarial modeling techniques can be connected to increasingly sophisticated AI architectures.

Xiaolin Zeng -- Susy hyperbolic sigma model on the two-dimensional hierarchical lattice

After introducing the random Schrodinger operator representation of the susy hyperbolic sigma H22 model on general graphs, we will discuss the special case of this model on the two-dimensional Dyson hierarchical lattice. We will explain how its exact fine-graining property can produce a limiting random measure on the unit interval, the latter can be called the Inverse Gaussian multiplicative chaos, and some of its properties are related to the properties of the H22 model. If time allows, we will explain how one can get optimal decay of the Green function of this model for all temperature on the 2d hierarchical lattice, a probe for the expected decay on Z^2.

Based on joint works of Yichao Huang and Jinglin Wang.

Martin Zerner – $p$-Fragility: Hanging by a Thousand Threads

 

 

 

◄ Back to: News & Events
To top