Hi! PARIS at ICML 2022

10 papers accepted at the 2022 International Conference on Machine Learning (ICML)

Hi! PARIS is pleased to present the work of its researchers and doctoral students at the 39th International Conference on Machine Learning (ICML) . The event will take place in Baltimore, Maryland, USA, from July 17 to 23, 2022.  

ICML is one of the most prestigious and longest-running general conference on machine learning and artificial intelligence. This world-class international gathering is renowned for its presentations and publications on cutting-edge research in all subjects of machine learning.  

A total of ten articles by Hi! PARIS afilliates were accepted after the demanding and thorough peer-review process for their significant methodological contribution to the field. This achievement recognizes the successful work of researchers from Hi! PARIS and renowned partner institutions such as Télécom Paris, École Polytechnique, Télécom SudParis and ENSAE Paris. 

Accepted papers will be published in the conference proceedings. These publications are a testament to the outstanding results of researchers in cutting-edge areas such as optimal transport, generative modeling and mathematical statistics. This international engagement will help to support the research activities of Hi! PARIS and promote French expertise in AI and machine learning on the international stage. 

We hope that in the years to come Hi! PARIS will have the means to develop an even more ambitious research agenda, from the mathematical foundations of machine learning to its applications, in order to gain a competitive edge and propel French AI research in the spotlight. 

Congratulations to our researchers!
For more details, see the list of publications and links below.

Congratulations to our researchers!

École polytechnique, IP PARIS: Aymeric Dieuleveut, Rémi Flamary, Eric Moulines, Charles Ollion, Guillaume Quispe, Erwan Scornet, Zoltan Szabo, Margaux Zaffran

ENSAE Paris, IP PARIS: Marco Cuturi, Anna Korba, Meyer Scetbon

Télécom Paris, IP PARIS: Florence d’Alché-Buc, Dimitri Bouche, Luc Brogat-Motte, Stephan Clémençon, Jean-Rémy Conti, Nathan Noiry

Télécom SudParis, IP PARIS: Max Cohen, Sylvain Le Corff

Here is the complete list of ICML publications for Hi! PARIS Research Affiliates:
  1. Adaptive Conformal Predictions for Time Series
    Margaux Zaffran (INRIA) · Olivier FERON (EDF) · Yannig Goude (EDF Lab Paris-Saclay) · julie Josse (Polytechnique/INRIA) · Aymeric Dieuleveut (École polytechnique)[ABSTRACT]

  2. Accurate Quantization of Measures via Interacting Particle-based Optimization
    Anna Korba (CREST/ENSAE) · Dejan Slepcev (Carnegie Mellon University) · Lantian Xu (Carnegie Mellon University) [ABSTRACT]

  3. An Asymptotic Test for Conditional Independence using Analytic Kernel Embeddings
    Meyer Scetbon (CREST, ENSAE) · Laurent Meunier (Dauphine University – FAIR Paris) · Yaniv Romano (Technion-Israel Institute of Technology)[ABSTRACT]

  4. Debiaser Beware: Pitfalls of Centering Regularized Transport Maps
    Aram-Alexandre Pooladian (New York University) · Jonathan Niles-Weed (NYU) · Marco Cuturi (ENSAE/CREST)[ABSTRACT]

  5. Diffusion bridges vector quantized variational autoencoders
    Max Cohen (Télécom SudParis) · Guillaume QUISPE (Polytechnique) · Sylvain Le Corff (Télécom SudParis) · Charles Ollion (Polytechnique, IPP) · Eric Moulines (Ecole Polytechnique)[ABSTRACT]

  6. Functional Output Regression with Infimal Convolution: Exploring the Huber and ϵϵ-insensitive Losses
    Alex Lambert (KU Leuven) · Dimitri Bouche (Télécom Paris) · Zoltan Szabo (Ecole Polytechnique) · Florence d’Alché-Buc (Télécom Paris, Institut Polytechnique de Paris)[ABSTRACT]

  7. Learning to Predict Graphs with Fused Gromov-Wasserstein Barycenters
    Luc Brogat-Motte (Télécom Paris) · Rémi Flamary (École Polytechnique) · Celine Brouard (INRAE) · Juho Rousu (Aalto University) · Florence d’Alché-Buc (Télécom Paris, Institut Polytechnique de Paris)[ABSTRACT]

  8. Linear-Time Gromov Wasserstein Distances using Low Rank Couplings and Costs
    Meyer Scetbon (CREST, ENSAE) · Gabriel Peyré (CNRS and ENS) · Marco Cuturi (ENSAE – IP PARIS / Google) [ABSTRACT]

  9. Mitigating Gender Bias in Face Recognition using the von Mises-Fisher Mixture Model
    Jean-Rémy Conti (Télécom Paris – IP PARIS / Idemia) · Nathan NOIRY (Télécom Paris – IP PARIS) · Vincent Despiegel (Idemia) · Stéphane Gentric (Idemia) · Stephan Clemencon (Télécom Paris – IP PARIS)[ABSTRACT]

  10. Near-optimal rate of consistency for linear models with missing values
    Alexis Ayme (Sorbonne Université) · Claire Boyer (LPSM, Sorbonne Université) · Aymeric Dieuleveut (École polytechnique) · Erwan Scornet (École Polytechnique)[ABSTRACT]