DATAIA Seminar | Marianne Clausel "Long range dependence in Machine Learning"

2024-03-21 12:30 2024-03-21 14:00 DATAIA Seminar | Marianne Clausel "Long range dependence in Machine Learning" Marianne Clausel (IECL, Université de Lorraine) has been a professor at the Université de Lorraine, affiliated with IECL, since September 2017. She is also an external collaborator of the SiMul team at CRAN. Title: Long range dependence in Machine Learning Summary: modeling long memory in data is a central challenge facing modern applications of machine learning to time series. Here we present a way of formalizing the concept of long memory. We explain how long memory can be estimated in practice, and study its impact in machine learning algorithms from both an empirical and a theoretical point of view. We present two applications: recommendation systems and sequence-to-sequence neural networks. The seminar will take place on Thursday, March 21, 2024, from 12:30 to 2:00 pm at CentraleSupélec, amphithéâtre e.068 (Bouygues building) in Gif s/Yvette.This seminar will also be broadcast by videoconference (link to come). CentraleSupélec, Amphithéâtre e.068 (bâtiment Bouygues), Gif-sur-Yvette

Thematic : Research

As part of its scientific activities, the DATAIA Institute organizes seminars throughout the year, with the aim of exchanging ideas on AI.

Public

Tout public
Event type

Conférence / séminaire / webinaire
Conditions

Registration required
Dates

Thursday 21 March, 12:30
12:30 pm - 02:00 pm
Location

CentraleSupélec, Amphithéâtre e.068 (bâtiment Bouygues), Gif-sur-Yvette

Marianne Clausel (IECL, Université de Lorraine) has been a professor at the Université de Lorraine, affiliated with IECL, since September 2017. She is also an external collaborator of the SiMul team at CRAN.

Title: Long range dependence in Machine Learning
Summary: modeling long memory in data is a central challenge facing modern applications of machine learning to time series. Here we present a way of formalizing the concept of long memory. We explain how long memory can be estimated in practice, and study its impact in machine learning algorithms from both an empirical and a theoretical point of view. We present two applications: recommendation systems and sequence-to-sequence neural networks.

The seminar will take place on Thursday, March 21, 2024, from 12:30 to 2:00 pm at CentraleSupélec, amphithéâtre e.068 (Bouygues building) in Gif s/Yvette.This seminar will also be broadcast by videoconference (link to come).