EPFL Topology Seminar 2023 Spring

 Location: Real World (And sometimes Zoom)

(For questions about the seminar, please contact the organizers:
Beware, there has been a change in rooms: new room is CO 122


Date Time Place Title Speaker
28.03.2023 14:15 CET MA A3 30 Relative plus construction Jérôme Scherer, EPFL
04.03.2023 14:15 CET MA A3 30 Hilton-Milnor’s theorem in infinity-toposes Samuel Lavenir, EPFL
11.04.2023     No seminar.  No seminar.
18.04.2023 14:15 CET MA A3 30 On deformations and lifts of Calabi-Yau varieties in characteristic p  

Lukas Brantner, University of Oxford and CNRS, Orsay

25.04.2023 14:15 CET CO 122 Integral representations of general linear groups and étale abelian motives Markus Spitzweck, Universität Osnabrück
02.05.2023 14:15 CET CO 122 Moduli spaces of some rational homotopy types Mario Fuentes Rumí, Universidad de Málaga
09.05.2023 14:15 CET CO 122 The multiplicativity of Euler characteristics and Becker-Gottlieb transfers Maxime Ramzi, University of Copenhagen
16.05.2023 14:15 CET CO 122 Abstract motivic homotopy theory Peter Arndt, University of Düsseldorf 
23.05.2023 14:15 CET CO 122

Weak omega-categories as a model for directed homotopy types

Hadrian Heine, EPFL
30.05.2023 14:15 CET CO 122

No seminar 

No seminar 
06.06.2023 14:15 CET CO 122

Persistence and the Sheaf-Function Correspondence

Nicolas Berkouk, EPFL
13.06.2023 14:15 CET CO 122


Jérôme Scherer,
Relative plus construction:

This is the project with Guille Carrion that we worked on during his stay in our group last Fall. A relative plus construction appears in a recent paper by Broto, Levi, and Oliver in the form of a classical Quillen cell attachment. The aim is to kill a given subgroup of the fundamental group without changing the homology with coefficients in a ring R. We provide a functorial version by working in the category of maps of spaces, which allows us to explain why the BLO version is not uniquely defined and why the kind of subgroup one can kill have different nature depending on the characteristic and the torsion in the ring R.



Samuel Lavenir,
Hilton-Milnor’s theorem in infinity-toposes:

Hilton-Milnor’s theorem is an important result in homotopy theory which provides a decomposition of the loop space on a finite wedge of suspension spaces into a certain (weak) infinite product. We have shown that a version of this theorem holds in any infinity-topos, and in hypercomplete infinity-categories with universal pushouts. This framework significantly simplifies the proof and allows to apply the theorem in a wide variety of contexts. These ideas build on recent work by Anel-Biedermann-Joyal-Finster and Devalapurkar-Haine on synthetic homotopy theory in higher toposes. After providing some context for Hilton-Milnor’s theorem, I will explain the main ideas of its higher categorical proof. If time permits, I will mention some possible applications in the theory of higher groups.



Lukas Brantner,
On deformations and lifts of Calabi-Yau varieties in characteristic p:

A smooth projective variety Z is said to be Calabi-Yau if its canonical bundle is trivial. I will discuss recent joint work with Taelman, in which we use derived algebraic geometry to study how Calabi-Yau varieties in characteristic p deform. More precisely, we show that if Z has degenerating Hodge–de Rham spectral sequence and torsion-free crystalline cohomology, then its mixed chracteristic deformations are unobstructed; this is an analogue of the classical BTT theorem in characteristic zero. If Z is ordinary, we show that it moreover admits a canonical lift to characteristic zero; this extends classical Serre-Tate theory. Our work generalises results of Achinger–Zdanowicz, Bogomolov-Tian-Todorov, Deligne–Nygaard, Ekedahl–Shepherd-Barron, Schröer, Serre–Tate, and Ward.


Markus Spitzweck,
Integral representations of general linear groups and étale abelian motives

After reviewing the envisioned Tannakian viewpoint on mixed motives and giving various definitions of motivic categories I will explain how one can model a subcategory of the category of étale motives generated by a 1-motive by modules over a commutative algebra in a stable category of representations of a general linear group over the integers. This builds upon rational results by Iwanari.

Mario Fuentes Rumí,
Moduli spaces of some rational homotopy types

Thanks to Daniel Quillen’s approach to rational homotopy theory, we can study topological spaces through their respective `models’ of Lie algebras. For example, if we fix a homology, we can ask which are the different homotopy types of topological spaces that share that homology (and similarly with homotopy groups or cohomology algebras). From the algebraic point of view, this question corresponds to finding the different differentials that we can place in a certain fixed Lie algebra $L$. This question induces a deformation problem and, following Deligne’s principle, these deformations will be given by the Maurer-Cartan elements in the derivations of L.

In this talk, we will introduce these concepts and study some simple properties of the moduli space associated with the Maurer-Cartan elements of complete Lie algebras of derivations.


Maxime Ramzi,

The multiplicativity of Euler characteristics and Becker-Gottlieb transfers: 

The Euler characteristic is a fundamental invariant of spaces and has many pleasant properties with respect to various ways of ‘decomposing’ spaces. One of these properties is its multiplicativity: if F-> E -> B is a fiber sequence of connected finite spaces, chi(E) = chi(F) chi(B).  A natural class of spaces for which the Euler characteristic still makes sense is the class of finitely dominated spaces – in this talk, I will discuss proofs of the corresponding equality when B,E and F are only assumed to be finitely dominated. I will explain the connection between this seemingly trivial (with nontrivial proof) property and deep questions in algebraic K-theory.

Parts of this talk are based on joint work with John Klein and Cary Malkiewich, and (time permitting) others on joint work with Shachar Carmeli, Bastiaan Cnossen and Lior Yanovski.



Peter Arndt,
Abstract motivic homotopy theory:

We will start with a quick introduction to motivic homotopy theory, emphasizing the parallels to the homotopy theory of topological spaces. We will then develop some basic motivic homotopy theory in an abstract setup: We replace the infinity category of motivic spaces by a presentable, cartesian closed infinity category (PCCC) and the multiplicative group scheme by a commutative group object G in this category. Starting from this, we will construct basic geometric objects like projective spaces, see a representation theorem for G-bundles, a Snaith type algebraic K-theory spectrum, Adams operations, rational splittings and a rational Eilenberg-MacLane spectrum. Examples of our abstract setup include motivic spaces built from algebraic geometry, complex and non-archimedian analytic geometry, derived geometry, log geometry, as well as the many proposals for geometry over the field with one element. Furthermore, there is an initial example of our setup, the classifying PCCC for commutative groups, which is a model for univalent homotopy type theory. Certain results and constructions, including all those from the talk, carry over from this particular PCCC with its generic commutative group to general PCCCs with commutative groups.


Hadrian Heine,

Weak omega-categories as a model for directed homotopy types

I will discuss a natural extension of homotopy theory to directed homotopy theory and explain why weak omega-categories are an effective model for directed homotopy types. I will show that many concepts of homotopy theory have directed analogues in the omega-categorical world. (This is joint work with David Gepner).


Nicolas Berkouk,
Persistence and the Sheaf-Function Correspondence:

The sheaf-function correspondence identifies the group of constructible functions on a real analytic manifold M with the Grothendieck group of constructible sheaves on M. When M is a finite dimensional real vector space, Kashiwara-Schapira have recently introduced the convolution distance between sheaves of k-vector spaces on M. In this paper, we characterize distances on the group of constructible functions on a real finite dimensional vector space that can be controlled by the convolution distance through the sheaf-function correspondence. Our main result asserts that such distances are almost trivial: they vanish as soon as two constructible functions have the same Euler integral. We formulate consequences of our result for Topological Data Analysis: there cannot exists non-trivial additive invariants of persistence modules that are continuous for the interleaving distance.