Seminars, for informal dissemination of research results, exploratory work by research teams, outreach activities, etc., constitute the simplest form of meetings at a Mathematics research centre.
CAMGSD has recorded the calendar of its seminars for a long time, this page serving both as a means of public announcement of forthcoming activities but also as a historic record.
The aim of this talk is to construct a measure-valued Markov process, representing an evolving population, such that its genealogies are given, in some convenient sense, by the Lambda-coalescents (Pitman 1999). Our starting point for this work is the celebrated duality between Fleming-Viot processes and coalescents (Perkins 1992. Bertoin and Le Gall 2003). Our motivation comes from several recent works establishing such connections when the evolving process has the branching property. We are able to extend this duality between two Markov additive processes, a forward one related to the Fleming-Viot process and a backward one related to the coalescent. After some transformation and time change of the forward process, we obtain a wide new class of measure-valued self-similar Markov processes, characterized by the self-similarity index and a Lévy triplet, that fulfills the desired conditions. This provides a change of paradigm in population genetics where the branching property is let aside. This is a joint work with Alejandro H. Wences.
A rigorous understanding of the dynamical nature of spacelike singularities remains an open problem in mathematical cosmology. Since the heuristic work of Belinski–Khalatnikov–Lifshitz and Misner's Mixmaster construction, vacuum spatially homogeneous cosmological models are expected to play a key role for generic singularities. We therefore focus on this class of models. The most general cases are the Bianchi type VIII, type IX, and type VI$_{-1/9}$, each with a four-dimensional Hubble-normalized state space.
On one hand, we embed the types VIII and IX models into modified gravity theories and show that general relativity (GR) arises as a bifurcation point where chaotic dynamics become generic, suggesting a new approximation scheme for GR. On the other hand, we analyze the type VI$_{-1/9}$ oscillatory regime and show that only a subset of its structure is dynamically relevant.
The Hopfield model stands as a paradigm at the intersection of statistical physics, theoretical neuroscience, and machine learning. Originally introduced as a biologically inspired model of associative memory, it has since evolved into a foundational framework for understanding a wide range of complex systems.
On the one hand, its roots in neuroscience enable a fruitful cross-fertilization: biologically grounded mechanisms continue to inspire algorithmic refinements and performance improvements in modern associative memory models. On the other hand, its formal connection with Boltzmann machines provides a bridge to contemporary machine learning techniques, including strategies such as dropout, pre-training, and the optimization of activation functions.
From the perspective of statistical mechanics, the Hopfield model remains a cornerstone for the analytical study of high-dimensional systems with disorder and frustration. This viewpoint naturally extends to the investigation of structured datasets, where the model offers a tractable yet expressive starting point for developing analytical insights.
In this talk, after a gentle introduction to the model, we will highlight some of these current research directions, while keeping the presentation accessible to a non-technical audience.
