Benjamin Heymann
I am a Staff Research Scientist at Criteo AI Lab, where I am the lead of the Performance Science team. I am a member of the join Inria/Criteo FairPlay team. I have a focus on game theory. I dedicate time to both applied and theoretical topics.
I hold a degree in engineering from the Ecole polytechnique (Paris), a Master of Science in Operations Research from Columbia University (New York) and a PhD in applied mathematics from the Ecole polytechnique/Inria.
I am currently co-advising Mélissa Tamine.
I was honoured to receive a Siebel Scholarship in 2016.
news
| Sep 05, 2025 | New extended abstracts for Consequences’25 : Non-Linear Counterfactual Aggregate Optimization with Otmane Sakhi , Confounding is a Pervasive Problem in Real World Recommender Systems with Alexander Merkov, David Rohde and Alexandre Gilotte, Adaptive Preference Aggregation for Recommender Systems, and Offline Contextual Bandit with Counterfactual Sample Identification with Alexandre Gilotte, Otmane Sakhi , Imad Aouali. |
|---|---|
| Sep 05, 2025 | New preprint: Counterfactual simulations for large scale systems with burnout variables |
| Sep 03, 2025 | My team got 5 papers accepted at Consequences’25 !!! |
| Aug 01, 2025 | I presented Learning in Games with progressive hiding (with Marc Lanctot), and Optimality Gap Analysis of the Decomposition-Coordination Method for Finite Horizon Bandit Problem (with Michel De Lara and Jean-Philippe Chancelier) at ICSP |
| Mar 10, 2025 | Adaptive Preference Aggregation has been accepted to GAIW at AAMAS 2025 !!! (thank you Marc Lanctot for your insights) |
| Feb 10, 2025 | On the Impact of the Utility in Semivalue-based Data Valuation with Mélissa Tamine, Patrick Loiseau and Maxime Vono is on arxiv. |
| Dec 20, 2024 | Our paper Learning in Games with progressive hiding (with Marc Lanctot) has been accepted for AAMAS 2025 !!! |
| Sep 25, 2024 | Our paper DU-Shapley: A Shapley Value Proxy for Efficient Dataset Valuation (with Felipe Garrido-Lucero, Maxime Vono, Patrick Loiseau, Vianney Perchet) has been accepted at NEURIPS 2024. |
| Sep 25, 2024 | Our paper Learning in Games with progressive hiding (with Marc Lanctot) has been selected for an oral presentation at Montreal AI symposium |
| Sep 16, 2024 | Our paper Repeated Bidding with Dynamic Value is accepted to WINE 2024 (with Alexandre Gilotte and Rémi Chan-Renous) |
| Sep 12, 2024 | Our paper Welfare-Optimized Recommender Systems is accepted for the workshop on Strategic and Utility-aware REcommendation (SURE) at the 18th ACM Conference on Recommender Systems (RecSys24)! |
| Sep 05, 2024 | Our preprint Learning in Games with progressive hiding is on arxiv (with Marc Lanctot). |
| Jul 15, 2024 | New preprint: A pragmatic policy learning approach to account for users’ fatigue in repeated auctions, with Alexandre Gilotte and Rémi Chan-Renous |
| Jul 10, 2024 | Our article Procurement Auctions with Losses with Alejandro Jofré has been published, here is a temporary link: here |
| May 07, 2024 | Updated preprint on Lagrangian relaxation and dynamic programming: Optimality Gap Analysis of the Decomposition-Coordination Method for Finite Horizon Bandit Problem. |
| Apr 04, 2024 | Our article Procurement Auctions with Losses with Alejandro Jofré has been accepted for publication at Computational Management Science!!! |
| Mar 28, 2024 | Our preprint Conditional Separation as a Binary Relation. A Coq Assisted Proof, with Jean-Philippe Chancelier and Michel De Lara, is available on HAL. |
| Mar 27, 2024 | Our paper Fixed Point Label Attribution for Real-Time Bidding, with Antoine Désir and Martin Bompaire, is published !!! |
| Mar 01, 2024 | Poster for Maximizing the Success Probability of Policy Allocations in Online Systems presented at AAAI: here. |
| Feb 22, 2024 | Today, I am presenting our poster about Maximizing the Success Probability of Policy Allocations in Online Systems at AAAI. |
| Feb 06, 2024 | I am glad to share our recently published article with Alejandro Jofré on reverse auctions with transportation. The link is here. |
| Jan 31, 2024 | Fixed point label attribution for real-time bidding, with Antoine Désir and Martin Bompaire has been accepted for publication at MSOM ! |
| Dec 20, 2023 | Our paper Reverse auctions with transportation and convex costs was accepted for publication in Computational Management Science (with Alejandro Jofré) |
| Dec 19, 2023 | Our paper Maximizing the Success Probability of Policy Allocations in Online Systems was accepted to AAAI-24 (with Artem Betlei, Thibaud Rahier, Mariia Vladimirova, Mehdi Sebbar and Nicolas Urien) |
| Nov 29, 2023 | My colleagues Jean-Philippe Chancelier and Michel De Lara presented our work at the Workshop on restless bandits, index policies and applications in reinforcement learning (slides available on the workshop page) |
| Oct 15, 2023 | Talk at Causality Discussion Group (online): Causality with Information Algebras. Presentation here. Thank Matej for the invitation!. |
| Oct 01, 2023 | New blog post with Flavian Vasile on Medium : Recommender systems need a user model |
| Oct 01, 2023 | An old work now on arxiv : Some bidding games converging to their unique pure equilibrium |
| Sep 01, 2023 | I will present at Journée des rencontres ENSAE-ENSAI our work on repeated bidding, presentation here, preprint here |
| Jul 01, 2023 | Updated preprint on learning in display advertising auctions Fixed point label attribution for real-time bidding. Joint work with Antoine Désir (INSEAD) and Martin Bompaire (Criteo AI Lab). |
| Jul 01, 2023 | New preprint on bidding in repeated auctions Repeated Bidding with Dynamic Value. Joint work with Alexandre Gilotte and Rémi Chan-Renous. |
| Jun 01, 2023 | I will be a speaker at Causality in Practice. Our presentation decks are here and here . |
| Jun 01, 2023 | New preprint on data sharing DU-Shapley: A Shapley Value Proxy for Efficient Dataset Valuation. |
| May 01, 2023 | We will present at Fundamental Challenges in Causality our work on Causal Inference with Information Fields. |
| Apr 01, 2023 | I will present at The Colloquium on When Causal Inference meets Statistical Analysis our work on Causal Inference with Information Fields (with Michel De Lara and Jean-Philippe Chancelier). A full version of the investigation is on arxiv (splitted in three subpapers). You can find the slides of my presentation here. |