Research

Hi! PARIS at NeurIPS 2021

24 research papers accepted at the conference on Neural Information Processing Systems (NeurIPS) 2021

Hi! PARIS is pleased to present the work of its researchers and professors at the 35th Conference on Neural Information Processing Systems (NeurIPS) being held this week (Dec 6-14, 2021).

For this edition, 24 publications were accepted by the reviewers for their significant contribution in the field of machine learning. The NeurIPS conference is world-renowned as one of the largest and most important machine learning conferences in the world.

The accepted papers cover a range of topics of AI (missing values prediction, Optimization under adversarial noise, reinforcement learning, interpretable machine learning, compressibility of overparametrized neural networks, federated learning, neuroimaging, ….). HEC Paris and several schools of Institut Polytechnique de Paris are represented.

Congratulations to our Professors and Researchers!
École polytechnique, Institut Polytechnique de Paris: Aymeric Dieuleveut, Rémi Flamary, Eric Moulines, Charles Ollion, Constantin Philippenko, Erwan Scornet, Achille Thin ENSAE Paris, Institut Polytechnique de Paris: Arya Akhavan, Etienne Boursier, Cristina Butucea, Evrard Garcelon, Yann Issartel, Olga Klopp, Vianney Perchet, Flore Sentenac, Alexandre Tsybakov HEC Paris: Julien Grand-Clément Inria Saclay: Alexandre Gramfort, Federica Granese, Pedro L. C. Rodrigues, Marine Le Morvan, Thomas Moreau, Catuscia Palamidessi, Hugo Richard, Bertrand Thirion, Gael Varoquaux Telecom Paris, Institut Polytechnique de Paris: Florence D’Alché-Buc, Kamélia Daudel, Pavlo Mozharovskyi, Nathan Noiry, Jayneel Parekh, Gaël Richard, Milad Sefidgaran Telecom Sud Paris, Institut Polytechnique de Paris: Randal Douc, Yazid Janati, Sylvain Le Corff
Here is the complete list of NeurIPS publications for Professors and Researchers in Hi! PARIS:
  1. DOCTOR: A Simple Method for Detecting Misclassification Errors
    Federica Granese, Marco Romanelli, Daniele Gorla, Catuscia Palamidessi, Pablo Piantanida [link] SPOTLIGHT
    Article available on https://proceedings.neurips.cc/paper/2021/file/2cb6b10338a7fc4117a80da24b582060-Paper.pdf
  2. What’s a good imputation to predict with missing values?
    Marine Le Morvan, Julie Josse, Erwan Scornet, Gael Varoquaux [link] SPOTLIGHT
    Article available on https://proceedings.neurips.cc/paper/2021/file/5fe8fdc79ce292c39c5f209d734b7206-Paper.pdf
  3. Decentralized Learning in Online Queuing Systems
    Flore Sentenac, Etienne Boursier, Vianney Perchet [link] SPOTLIGHT
    Article available on https://proceedings.neurips.cc/paper/2021/file/99ef04eb612baf0e86671a5109e22154-Paper.pdf
  4. Online Sign Identification: Minimization of the Number of Errors in Thresholding Bandits
    Reda Ouhamma, Rémy Degenne, Vianney Perchet, Pierre Gaillard [link] SPOTLIGHT
    Article available on https://papers.neurips.cc/paper/2021/file/9a3f54913bf27e648d1759c18d007165-Paper.pdf
  5. Disentangling Identifiable Features from Noisy Data with Structured Nonlinear ICA
    Hermanni Hälvä, Sylvain Le Corff, Luc Lehéricy, Jonathan So, Yongjie Zhu, Elisabeth Gassiat, Aapo Hyvarinen [link]
    Article available on https://arxiv.org/pdf/2106.09620.pdf
  6. Preserved central model for faster bidirectional compression in distributed settings
    Constantin Philippenko, Aymeric Dieuleveut [link]
    Article available on https://papers.nips.cc/paper/2021/file/13d63838ef1fb6f34ca2dc6821c60e49-Paper.pdf
  7. A New Theoretical Framework for Fast and Accurate Online Decision-Making
    Nicolò Cesa-Bianchi, Tom Cesari, Yishay Mansour, Vianney Perchet [link]
    Article available on https://papers.nips.cc/paper/2021/file/4c4ea5258ef3fb3fb1fc48fee9b4408c-Paper.pdf
  8. Conic Blackwell Algorithm: Parameter-Free Convex-Concave Saddle-Point Solving
    Julien Grand-Clément, Christian Kroer [link]
    Article available on https://papers.nips.cc/paper/2021/file/4f87658ef0de194413056248a00ce009-Paper.pdf
  9. Distributed Zero-Order Optimization under Adversarial Noise
    Arya Akhavan, Massimiliano Pontil, Alexandre Tsybakov [link]
    Article available on https://papers.nips.cc/paper/2021/file/5487e79fa0ccd0b79e5d4a4c8ced005d-Paper.pdf
  10. Local Differential Privacy for Regret Minimization in Reinforcement Learning
    Evrard Garcelon, Vianney Perchet, Ciara Pike-Burke, Matteo Pirotta [link]
    Article available on https://papers.nips.cc/paper/2021/file/580760fb5def6e2ca8eaf601236d5b08-Paper.pdf
  11. Making the most of your day: online learning for optimal allocation of time
    Etienne Boursier, Tristan Garrec, Vianney Perchet, Marco Scarsini [link]
    Article available on https://papers.nips.cc/paper/2021/file/5d2c2cee8ab0b9a36bd1ed7196bd6c4a-Supplemental.pdf
  12. NEO: Non Equilibrium Sampling on the Orbits of a Deterministic Transform
    Achille Thin, Yazid Janati El Idrissi, Sylvain Le Corff, Charles Ollion, Eric Moulines, Arnaud Doucet, Alain Durmus, Christian Robert [link]
    Article available on https://papers.nips.cc/paper/2021/file/8dd291cbea8f231982db0fb1716dfc55-Paper.pdf
  13. Optimality of variational inference for stochasticblock model with missing links
    Solenne Gaucher, Olga Klopp [link]
    Article available on https://papers.nips.cc/paper/2021/file/a5e308070bd6dd3cc56283f2313522de-Paper.pdf
  14. Online Matching in Sparse Random Graphs: Non-Asymptotic Performances of Greedy Algorithm
    Nathan Noiry, Vianney Perchet, Flore Sentenac [link]
    Article available on https://papers.nips.cc/paper/2021/file/b294504229c668e750dfcc4ea9617f0a-Supplemental.pdf
  15. Unbalanced Optimal Transport through Non-negative Penalized Linear Regression
    Laetitia Chapel, Rémi Flamary, Haoran Wu, Cédric Févotte, Gilles Gasso [link]
    Article available on https://papers.neurips.cc/paper/2021/file/c3c617a9b80b3ae1ebd868b0017cc349-Paper.pdf
  16. A Framework to Learn with Interpretation
    Jayneel Parekh, Pavlo Mozharovskyi, Florence D’Alché-Buc [link]
    Article available on https://proceedings.neurips.cc/paper/2021/file/cbb6a3b884f4f88b3a8e3d44c636cbd8-Paper.pdf
  17. Stochastic Online Linear Regression: the Forward Algorithm to Replace RidgeReda Ouhamma, Odalric-Ambrym Maillard, Vianney Perchet [link]
    Article available on https://proceedings.neurips.cc/paper/2021/file/cca289d2a4acd14c1cd9a84ffb41dd29-Paper.pdf
  18. Locally differentially private estimation of nonlinear functionals of discrete distributions
    Cristina Butucea, Yann Issartel [link]
    Article available on https://papers.nips.cc/paper/2021/file/cf8c9be2a4508a24ae92c9d3d379131d-Paper.pdf
  19. Heavy Tails in SGD and Compressibility of Overparametrized Neural Networks
    Melih Barsbey, Seyedmilad Sefidgaran, Murat A. Erdogdu, Gaël Richard, Umut Simsekli [link]
    Article available on https://proceedings.neurips.cc/paper/2021/file/f5c3dd7514bf620a1b85450d2ae374b1-Paper.pdf
  20. Federated-EM with heterogeneity mitigation and variance reduction
    Aymeric Dieuleveut, Gersende Fort, Eric Moulines, Geneviève Robin [link]
    Article available on https://proceedings.neurips.cc/paper/2021/file/f740c8d9c193f16d8a07d3a8a751d13f-Paper.pdf
  21. Tight High Probability Bounds for Linear Stochastic Approximation with Fixed Stepsize
    Alain Durmus, Eric Moulines, Alexey Naumov, Sergey Samsonov, Kevin Scaman, Hoi-To Wai [link]
    Article available on https://papers.nips.cc/paper/2021/file/fc95fa5740ba01a870cfa52f671fe1e4-Paper.pdf
  22. Shared Independent Component Analysis for Multi-Subject Neuroimaging
    Hugo Richard, Pierre Ablin, Bertrand Thirion, Alexandre Gramfort, Aapo Hyvarinen [link]
    Article available on https://proceedings.neurips.cc/paper/2021/file/fb508ef074ee78a0e58c68be06d8a2eb-Paper.pdf
  23. HNPE: Leveraging Global Parameters for Neural Posterior Estimation
    Pedro Luiz Coelho Rodrigues, Thomas Moreau, Gilles Louppe, Alexandre Gramfort [link]
    Article available on https://hal.archives-ouvertes.fr/hal-03139916/document
  24. Mixture weights optimisation for Alpha-Divergence Variational Inference
    Kamélia Daudel, Randal Douc [link]
    Article available on https://papers.nips.cc/paper/2021/file/233f1dd0f3f537bcb7a338ea74d63483-Paper.pdf