Subspace arrangements and the representation theory of rational Cherednik algebras
Abstract
I will explain how the representation theory of rational Cherednik algebras interacts with the commutative algebra of certain subspace arrangements arising from the reflection arrangement of a complex reflection group. Potentially, the representation theory allows one to study both qualitative questions (e.g., is the arrangement Cohen-Macaulay or not?) and quantitative questions (e.g., what is the Hilbert series of the ideal of the arrangement, or even, what are its graded Betti numbers?), by applying the tools (such as orthogonal polynomials, Kazhdan-Lusztig characters, and Dirac cohomology) that representation theory provides. This talk is partly based on joint work with Susanna Fishel and Elizabeth Manosalva.
Stochastic Ricci flow on surfaces
Abstract
The Ricci flow on a surface is an intrinsic evolution of the metric converging to a constant curvature metric within the conformal class. It can be seen as an infinite-dimensional gradient flow. We introduce a natural 'Langevin' version of that flow, thus constructing an SPDE with invariant measure expressed in terms of Liouville Conformal Field Theory.
Joint work with Hao Shen (Wisconsin).
What is a unipotent representation?
Abstract
Let $G$ be a connected reductive algebraic group, and let $G(\mathbb{F}_q)$ be its group of $\mathbb{F}_q$-rational points. Denote by $\mathrm{Irr}(G(\mathbb{F}_q))$ the set of (equivalence classes) of irreducible finite-dimensional representations. Deligne and Lusztig defined a finite subset $$\mathrm{Unip}(G(\mathbb{F}_q)) \subset \mathrm{Irr}_{\mathrm{fd}}(G(\mathbb{F}_q))$$
of unipotent representations. These representations play a distinguished role in the representation theory of $G(\mathbb{F}_q)$. In particular, the classification of $\mathrm{Irr}_{\mathrm{fd}}(G(\mathbb{F}_q))$ reduces to the classification of $\mathrm{Unip}(G(\mathbb{F}_q))$.
Now replace $\mathbb{F}_q$ with a local field $k$ and replace $\mathrm{Irr}_{\mathrm{fd}}(G(\mathbb{F}_q))$ with $\mathrm{Irr}_{\mathrm{u}}(G(k))$ (irreducible unitary representations). Vogan has predicted the existence of a finite subset
$$\mathrm{Unip}(G(k)) \subset \mathrm{Irr}_{\mathrm{u}}(G(k))$$
which completes the following analogy
$$\mathrm{Unip}(G(k)) \text{ is to } \mathrm{Irr}_{\mathrm{u}}(G(k)) \text{ as } \mathrm{Unip}(G(\mathbb{F}_q)) \text{ is to } \mathrm{Irr}_{\mathrm{fd}}(G(\mathbb{F}_q)).$$
In this talk I will propose a definition of $\mathrm{Unip}(G(k))$ when $k = \mathbb{C}$. The definition is geometric and case-free. The representations considered include all of Arthur's, but also many others. After sketching the definition and cataloging its properties, I will explain a classification of $\mathrm{Unip}(G(\mathbb{C}))$, generalizing the well-known result of Barbasch-Vogan for Arthur's representations. Time permitting, I will discuss some speculations about the case of $k=\mathbb{R}$.
This talk is based on forthcoming joint work with Ivan Loseu and Dmitryo Matvieievskyi.
The Dirac inequality, Weyl groups, and isolated unitary representations
Abstract
In the classical setting of real semisimple Lie groups, the Dirac inequality (due to Parthasarathy) gives a necessary condition that the infinitesimal character of an irreducible unitary representation needs to satisfy in terms of the restriction of the representation to the maximal compact subgroup. A similar tool was introduced in the setting of representations of p-adic groups in joint work with Barbasch and Trapa, where the necessary unitarity condition is phrased in terms of the semisimple parameter in the Kazhdan-Lusztig parameterization and the hyperspecial parahoric restriction. I will present several consequences of this inequality to the problem of understanding the unitary dual of the p-adic group, in particular, how it can be used in order to exhibit several isolated "extremal" unitary representations and to compute precise "spectral gaps" for them.
Classifier-based Distribution-Dissimilarities: From Maximum Mean Discrepancies to Adversarial Examples
datasig.ox.ac.uk/events
Abstract
Any binary classifier (or score-function) can be used to define a dissimilarity between two distributions of points with positive and negative labels. Actually, many well-known distribution-dissimilarities are classifier-based dissimilarities: the total variation, the KL- or JS-divergence, the Hellinger distance, etc. And many recent popular generative modelling algorithms compute or approximate these distribution-dissimilarities by explicitly training a classifier: eg GANs and their variants. After a brief introduction to these classifier-based dissimilarities, I will focus on the influence of the classifier's capacity. I will start with some theoretical considerations illustrated on maximum mean discrepancies --a weak form of total variation that has grown popular in machine learning-- and then focus on deep feed-forward networks and their vulnerability to adversarial examples. We will see that this vulnerability is already rooted in the design and capacity of our current networks, and will discuss ideas to tackle this vulnerability in future.
Atomic subgraphs and the statistical mechanics of networks
Abstract
We develop random graph models where graphs are generated by connecting not only pairs of vertices by edges but also larger subsets of vertices by copies of small atomic subgraphs of arbitrary topology. This allows the for the generation of graphs with extensive numbers of triangles and other network motifs commonly observed in many real world networks. More specifically we focus on maximum entropy ensembles under constraints placed on the counts and distributions of atomic subgraphs and derive general expressions for the entropy of such models. We also present a procedure for combining distributions of multiple atomic subgraphs that enables the construction of models with fewer parameters. Expanding the model to include atoms with edge and vertex labels we obtain a general class of models that can be parametrized in terms of basic building blocks and their distributions that includes many widely used models as special cases. These models include random graphs with arbitrary distributions of subgraphs, random hypergraphs, bipartite models, stochastic block models, models of multilayer networks and their degree corrected and directed versions. We show that the entropy for all these models can be derived from a single expression that is characterized by the symmetry groups of atomic subgraphs.
Model-independence in a fixed-income market and weak optimal transport
Abstract
In this talk I will consider model-independent pricing problems in a stochastic interest rates framework. In this case the usual tools from Optimal Transport and Skorokhod embedding cannot be applied. I will show how some pricing problems in a fixed-income market can be reformulated as Weak Optimal Transport (WOT) problems as introduced by Gozlan et al. I will present a super-replication theorem that follows from an extension of WOT results to the case of non-convex cost functions.
This talk is based on joint work with M. Beiglboeck and G. Pammer.