Mathieu Rita

Inria-Microsoft Research Joint Lab, Ecole Normale Supérieure.

profile_pic.jpeg

Centre Sciences des Données (CSD)

45 rue d'Ulm

75005, Paris, France

I am a Ph.D. student under the supervision of Emmanuel Dupoux (ENS/FAIR), Olivier Pietquin (Google Brain) and Florian Strub (DeepMind). I work between the Inria-Microsoft Research Joint Lab and the CoML team in Paris, which is located at ENS Paris.

Prior to that, I received an engineering degree from Ecole Polytechnique and a MSc degree in Mathematics, Computer Vision and Machine Learning from Ecole Normale Supérieure Paris-Saclay.

My research explores the theoretical and experimental aspects of training RL objectives with language models, with a specific focus on constructing self-play multi-agent systems. I particularly investigate how scaling populations and generations of agents can help address language learning challenges, such as overfitting, exploration or drift. As an application, I simulate language evolution and study the pre-requisites necessary to the emergence of language universals, such as compositionality.

News

Jan 22, 2023 [📝 PAPER] Our paper Revisiting Populations in Multi-Agent Communication has been accepted at ICLR’23.
Sep 14, 2022 [📝 PAPER] Our paper Emergent Communication: Generalization and Overfitting in Lewis Games has been accepted at NeurIPS’22.
Jul 3, 2022 [🇯🇵 WORKSHOP] Our Workshop “Machine Learning and the Evolution of Language” will take place at JCoLE’22 on September, 5th. Find all information on our webpage.
Apr 29, 2022 [💻 WORKSHOP] Our Workshop ‘Emergent communication: New frontiers’ takes place at ICLR’22 today
Apr 22, 2022 [📝 PAPER] Our paper On the Role of Population Heterogeneity in Emergent Communication is presented at ICLR’22 this week.
Feb 1, 2021 [✒️ PhD] I am starting a PhD on Emergent Communication under the supervision of Emmanuel Dupoux, Olivier Pietquin & Florian Strub.
Jan 15, 2021 [👨‍🎨 ART] Our generated video Dreamy Cops is exposed in the online CVPR Computer Vision Art Gallery.
Nov 20, 2020 [📝 PAPER] Our paper “LazImpa”: Lazy and Impatient neural agents learn to communicate efficiently is presented to CoNLL’20.

Selected publications

  1. Revisiting Populations in Multi-Agent Communication
    Michel, Paul, Rita, Mathieu, Mathewson, Kory Wallace, Tieleman, Olivier, and Lazaridou, Angeliki
    The 11th International Conference on Learning Representations (ICLR) 2023 2023
  2. Emergent Communication: Generalization and Overfitting in Lewis Games
    Rita, Mathieu, Tallec, Corentin, Michel, Paul, Grill, Jean-Bastien, Pietquin, Olivier, Dupoux, Emmanuel, and Strub, Florian
    36th Conference on Neural Information Processing Systems (NeurIPS) 2022
  3. On the role of population heterogeneity in emergent communication
    Rita, Mathieu, Strub, Florian, Grill, Jean-Bastien, Pietquin, Olivier, and Dupoux, Emmanuel
    The Tenth International Conference on Learning Representations (ICLR) 2022
  4. " LazImpa": Lazy and Impatient neural agents learn to communicate efficiently
    Rita, Mathieu, Chaabouni, Rahma, and Dupoux, Emmanuel
    In Proceedings of CoNLL 2020