EP-IT Data Science Seminars

Problem solving as a translation task

by Francois Charton (META AI)

503/1-001 - Council Chamber (CERN)

503/1-001 - Council Chamber


Show room on map
Neural architectures designed for machine translation can be used to solve problems of mathematics, by considering that solving amounts to translating the problem, a sentence in some mathematical language, into its solution, another sentence in mathematical language. Presenting examples from symbolic and numerical mathematics, and theoretical physics, I show how such techniques can be applied to develop AI for Science, and help understand the inner workings of language models.
Francois Charton is a research engineer in Meta AI,  working on applications of language models to mathematics and science. He is a former student of Ecole Polytechnique, and ENSAE (Ecole Nationale de la Statistique et de l'Administration Economique).  He has been in Meta since 2019, and had a career in industry prior to that.

Coffee will be served at 10:30.

Organized by

M. Girone, M. Elsing, L. Moneta, M. Pierini
Event co-organised with IML coordinators as part of the 6th IML Workshop (https://indico.cern.ch/event/1297159/)

There is a live webcast for this event