NeurIPS 2024: 19 papers selected at Inria Saclay Centre

Add alternative text: Neural Information Processing Systems

NeurIPS  is one of the most renowned international scientific conferences in the field of AI and machine learning. Founded in 1987 in Denver, U.S.A., every year it brings together researchers from all over the world working in the field.

The 6-day program includes keynotes from influential figures in the field, oral presentations and poster sessions, tutorials and workshops, competitions, demonstrations and community events.

A rigorous selection of accepted items

Each article submitted is evaluated by a committee of experts. These experts select only the most innovative contributions, with extremely rigorous methodology and careful consideration of the societal impact of the research. Since 2020, authors have also had to address the provision of data and code, ethical considerations and the limits of the work.

This year, out of over 15,000 papers submitted, less than 24% were accepted at NeurIPS 2024. And of these, 19 came from researchers who are members of an Inria Saclay Centre projects-teams.

This demanding approach maintains the NeurIPS conference as a global benchmark in the field.

A key meeting place for researchers

Beyond its scientific aspect, for our researchers selected to present their work, NeurIPS also represents an exceptional audience for their research.

The conference is also a key meeting place for academic researchers, industry and technology players, providing a forum for listening to each other's ideas and providing inspiration.

Attendees, who number in the thousands - around 10,000 for previous editions - discover the latest fundamental advances in AI that address concrete and urgent challenges in whole sectors such as health, the environment and transport.

Congratulations to our researchers, their co-authors and their partners for this success.

19 selected articles reflecting the diversity of our research 

• Celeste project-team (Inria, Université Paris-Saclay), LMO laboratory
  • Regression under demographic parity constraints via unlabeled post-processing, Evgenii Chzhen, Mohamed Hebiri, Gayane Taturyan
  • Unravelling in Collaborative Learning, Aymeric Capitaine, Etienne Boursier, Antoine Scheid, Eric Moulines, Michael Jordan, El-Mahdi El-Mhamdi, Alain Durmus
  • Learning to Mitigate Externalities: the Coase Theorem with Hindsight Rationality, Antoine Scheid, Aymeric Capitaine, Etienne Boursier, Eric Moulines, Michael Jordan, Alain Durmus
  • Addressing bias in online selection with limited budget of comparisons, Ziyad Benomar, Evgenii Chzhen, Nicolas Schreuder, Vianney Perchet
• DataShape project-team(Inria, Université Paris-Saclay), LMO laboratory
  • Wasserstein convergence of Čech persistence diagrams for samplings of submanifolds, Charles Arnal, David Cohen-Steiner, Vincent Divol 
  • Iteration Head: A Mechanistic Study of Chain-of-Thought, Vivien Cabannes, Charles Arnal, Wassim Bouaziz, Alice Yang, Francois Charton, Julia Kempe
  • Diffeomorphic interpolation for efficient persistence-based topological optimization, Mathieu Carriere, Marc Theveneau, Théo Lacombe
• FairPlay project-team(Inria, ENSAE Paris, Critéo), CREST laboratory 
  • Adressing bias in online selection wit limited budget of comparisons, Ziyad Benomar, Evgenii Chzhen, Nicolas Schreuder, Vianney Perchet
  • The Value of Reward Lookahed in Reinforcement Learning, Nadav Merlis, Dorian Baudry, Vianney Perchet 
  • Local and adaptive mirror descents in extensive-form games, Côme Fiegel, Pierre Menard, Tadashi Kozuno, Remi Munos, Vianney Perchet, Michal Valko
  • Improved Learning rates in multi-unit uniform price auctions, Marius Potfer, Dorian Baudry, Hugo Richard, Vianney Perchet, Cheng Wan
  • Lookback Prophet Inequalitites, Ziyad Benomar, Dorian Baudry, Vianney Perchet
  • Strategic Multi-Armed Bandit Problems Under Debt-Free Reporting, Ahmed Ben Yahmed, Clément Calauzènes, Vianney Perchet
  • Improved Algorithms for Contextual Dynamic Pricing, Matilde Tullii, Solenne Gaucher, Nadav Merlis, Vianney Perchet
  • Optimizing the coalition gain in Online Auctions with Greedy Structured Bandits, Dorian Baudry, Hugo Richard, Maria Cherifa, Vianney Perchet, Clément Calauzènes
  • DU-Shapley : A Shapley Value Proxy for Efficient dataset valuation, Felipe Garrido, Benjamin Heymann, Maxime Vono, Patrick Loiseau, Vianney Perchet
  • Reinforcement Learning with Lookahead information, Nadav Merlis
• GeomeriX project-team (Inria, École polytechnique, CNRS), LIX laboratory
  • DeBaRA: Denoising-Based 3D Room Arrangement Generation, Léopold Maillard, Nicolas Sereyjol-Garros, Tom Durand, Maks Ovsjanikov
• OPIS project-team (Inria, Université Paris-Saclay, CentraleSupélec), CVN laboratory
  • Continuous Product Graph Neural Networks, Aref Einizade, Fragkiskos D. Malliaros, and Jhony H. Giraldo
• Soda project-team (Inria)
  • Better by default: Strong pre-tuned MLPs and boosted trees on tabular data, David Holzmüller, Leo Grinsztajn, Ingo Steinwart