The Online Advertising industry is seeing a major shift today in its operational constraints with a global movement towards more privacy. Popular techniques for privacy-compliant advertising such as aggregation and differential privacy mechanisms were shown to match high privacy standards...
A Workshop organized by people from the Criteo AI Lab at KDD 2021 Summary. Machine learning has allowed many systems that we interact with to improve performance and personalize. An important source of information in these systems is to...
We are pleased to announce a privacy-preserving learning challenge at CAp’21. The goal is to find a representation of a stream of events (x) such that a prediction (y) task is possible while being able to detect that...
3 papers at @AISTATS 2021 from Criteo AI Lab folks Instance-Wise Minimax-Optimal Algorithms for Logistic Bandits. Marc Abeille, Louis Faury, Clément Calauzènes. Arxiv link. Self-Concordant Analysis of Generalized Linear Bandits with Forgetting. Yoan Russac (DI-ENS, CNRS, Inria, PSL, VALDA), Louis...
Happy to have 5 papers accepted at ICLR 2021 (among which 2 oral presentations) and 1 paper at ALT 2021! ICLR 2021: Scalable Learning and MAP Inference for Nonsymmetric Determinantal Point Processes (oral), Mike Gartrell, Insu Han, Elvis Dohmatob, Jennifer Gillenwater, Victor-Emmanuel...
We are happy to have 9 papers co-authored by researchers from the Criteo AI Lab at NeurIPS 2020 Online non-convex optimization with inexact models, A. Héliou, M. Martin, P. Mertikopoulos, and T. Rahier Explore aggressively, update conservatively: Stochastic extragradient...
Alex Gilotte, researcher at the Criteo AI Lab, details how counterfactual reasoning may be applied to evaluate a recommender system. The first posts introduce the topic and the intuitions behind the mathematics of counterfactual reasoning, and the later...
What defines large-scale machine learning? This seemingly innocent question is often answered with petabytes of data and hundreds of GPUs. It turns out that large-scale machine learning does not have much to do with all of that. In...
The paper on “A Principle of Least Action for the Training of Neural Networks” by Skander Karkar, Patrick Gallinari and their LIP6_lab co-authors, has been selected as the best student ML paper award Runner-up @ECMLPKDD! https://bit.ly/3gL19qA
The Reco team has released DeepR, a package to train deep recommendation models in production! See the full post here, with links to Github, docs, quick start guide and some examples: DeepR on Medium.