Neuromat Online Seminars Abstract

Mar 20, 2020

1. About
2. Poster
3. Abstracts
4. Presentation slides

April 23 Thursday at 10 AM BRT

Lecturers: Léo Planche and Morgan André

Affiliation: Research, Innovation and Dissemination Center for Neuromathematics (RIDC NeuroMat) doctoral researchers.

Title: "What's Metastability (in neural networks)?"

Abstract: The topic of metastability in the brain, or more generally in neural networks, has been increasingly discussed in the neuroscience litterature during the last decades. The exact meaning in which this word is used by the different authors of this litterature seems however unclear. On the other hand metastability is now a pretty well defined phenomenon in the fields of statistical physics and stochastic processes. In this talk we try to clarify what metastability in the brain could mean, and we show how the statistical physics definition of metastability can apply to biological neural networks by summarizing a few results of metastability obtained for a class of models inspired by the Galves-Löcherbach model.

---

May 7 Thursday at 10 AM BRT

Lecturers: Rafael B. Stern

Affiliation: Universidade Federal de São Carlos and Research, Innovation and Dissemination Center for Neuromathematics (RIDC NeuroMat) associate investigator.

Title: "Retrieving information from complex data: classifying Parkinson-related variables using the Goal Keeper Game"

Abstract: People with Parkinson's disease (PD) display poorer gait performance when walking under complex conditions than under simple conditions. Screening tests that evaluate gait performance changes under complex walking conditions may be valuable tools for early intervention, especially if allowing for massive data collection. The Goalkeeper Game (GG) might allow such a massive collection. In this presentation I discuss how to extract information from the GG that is useful for predicting gait performance.

References: Goalkeeper Game: A New Assessment Tool for Prediction of Gait Performance Under Complex Condition in People With Parkinson's Disease

---

May 21 Thursday at 10 AM BRT

Lecturers: Claudia D. Vargas, Oswaldo Baffa Filho and Risto Ilmoniemi

Affiliation: Instituto de Biofísica Carlos Chagas Filho – Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto – Aalto University. The three are also Research, Innovation and Dissemination Center for Neuromathematics (RIDC NeuroMat) co-principal investigators.

Title: "Structural learning and the brain: new conceptual and methodological venues"

Abstract: In this seminar, Claudia Vargas will discuss a new protocol to stimulate the motor area of the brain with sequences of transcranial magnetic stimulation pulses driven by stochastic chains of variable length and collect motor evoked potentials in muscles of interest. To perform this experimental protocol with increasing precision a neuronavigated TMS will be used and the InVesalius neuronavigator will be described by Oswaldo Baffa, as well as its future development with a collaborative robot (cobot) arm. Finally Risto Ilmoniemi will discuss the use of EEG/TMS techniques as well as a new multilocus TMS stimulator that can stimulate different areas of the brain within a millisecond time interval.

---

June 4 Thursday at 10 AM BRT

Lecturers: Guilherme Ost

Affiliation: Universidade Federal do Rio de Janeiro and Research, Innovation and Dissemination Center for Neuromathematics (RIDC NeuroMat) associate investigator.

Title: "Sparse space-time models"

Abstract: Inspired by Kalikow-type decompositions, we introduce a new stochastic model of infinite neuronal networks, for which we establish sharp oracle inequalities for Lasso methods and restricted eigenvalue properties for the associated Gram matrix with high probability. These results hold even if the network is only partially observed. The main argument rely on the fact that concentration inequalities can easily be derived whenever the transition probabilities of the underlying process admit a sparse space-time representation. This is a joint work with Patricia Reynaud-Bouret (Université Côte D'Azur) which was done as part of the activities of the sister projects NeuroMat and NeuroMod. It can be dowloaded from https://arxiv.org/abs/1807.07615To appear in Annales de l'Institut Henri Poincaré (B) Probability and Statistics, 2020.

---

Share on Twitter Share on Facebook