Algorithms
-
The Grind Behind the Epiphany: A Short Story of a Research Project
From ideation to outcome, this is the story of a privacy-preserving research project. It tells how research can generate innovations but also joy and despair. Early 2024: The “Hammer” Phase Two research leads who pioneered Criteo’s early privacy initiatives, as part of the Criteo multi-year research program, introduced me to a challenge born from the Privacy…
-
Off-Policy Learning in Large Action Spaces
Optimization Matters More Than Estimation This Research Card shows that in large-scale recommendation systems, like those used at Criteo, focusing on how we optimize algorithms is more impactful than trying to perfectly estimate outcomes. By using simpler, more stable training objectives, we can achieve better click-through rates, faster experimentation, and more scalable performance across millions…
-
Pareto-Optimality, Smoothness, and Stochasticity in Learning-Augmented One-Max-Search
A rule for choosing when to sell/buy an asset when facing a sequence of prices This Research Card explains a rule for choosing when to sell/buy an asset when facing a sequence of prices. The key feature is that the algorithm has access to “predictions” on the optimal price. The objectives are to take “safe” decisions,…
-
Our Experience at ALT 2025
A Journey into Algorithmic Learning Theory Authors: Marc Abeille, Julien Zhou & Corentin Pla. The Algorithmic Learning Theory (ALT) 2025 conference, held in Milan at the end of February, was a significant event in the field of machine learning and theoretical computer science. This one-track conference brought together hundred researchers and practitioners to discuss the latest…
-
Advent of Code 2024 — Sharing our experiences
At Criteo, we were founded on technology, and technology is our passion. We enjoy challenges, so we were ready to participate in Advent of Code once again this year. Additionally, as we did in 2023, we decided to support the initiative by becoming a sponsor. Community building is at the core of our mission, so…
-
Research Card: Mode Identification with Partial Feedback
An adaptive algorithm that learns the best sequence of questions to identify as quickly as possible (in terms of questions) the most probable label for data points. Paper: Mode Identification with Partial Feedback Authors: Vivien Cabannes (Meta), Charles Arnal (Univ. Paris Saclay), V. Perchet (Ensae & Criteo) Category: Annotation Revue: COLT 2024 Why did we work…
-
Research Card: Fixed Point Label Attribution for Real-Time Bidding
Research Card on how to attribute positive outcomes to touchpoints in a sequence of displays leading to a conversion. Paper: Fixed Point Label Attribution for Real-Time BiddingAuthors: Martin Bompaire, Antoine Désir, Benjamin HeymannCategory: Operations ResearchRevue: Manufacturing and Service Operations Management Why did we work on this topic (the problem we want to solve)? In display advertising,…
-
Research Card: Strategic Arms with Side Communication Prevail Over Low-Regret MAB Algorithms
Paper: Strategic Arms with Side Communication Prevail Over Low-Regret MAB AlgorithmsAuthors: Ahmed Ben Yahmed, Clément Calauzènes, Vianney PerchetCategory: Learning Theory, Sequential LearningRevue: ICASSP 2024 Why did we work on this topic (the problem we want to solve)? We chose to investigate this topic because we aimed to address a fundamental problem within the strategic multi-armed bandit…
-
Research Card: Federated Wasserstein Distance
How to tackle the challenge of calculating the similarity between two sets of embeddings, each held by a different party, while respecting privacy restrictions? Paper: Federated Wasserstein DistanceAuthors: Alain Rakotomamonjy (Criteo AI Lab), Kimia Nadjahi (MIT), Liva Ralaivola (Criteo AI Lab).Category: Federated Wasserstein DistanceRevue: ICLR 2024 Why did we work on this topic, the problem…
-
Criteo WomenDoTechToo Paris Meetup — Takeaways
Our “Women do tech too” meetup event gives the floor to women working at Criteo so they can share their technical expertise. It is a unique opportunity to meet, network, and share insights with other passionate women from the tech field. Who can join? Basically, anyone who believes in supporting a more diverse and inclusive…