no code implementations • 22 Feb 2024 • Imad Aouali, Victor-Emmanuel Brunel, David Rohde, Anna Korba
In this framework, we propose sDM, a generic Bayesian approach designed for OPE and OPL, grounded in both algorithmic and theoretical foundations.
no code implementations • 3 Feb 2024 • David Rohde
Applied recommender systems research is in a curious position.
no code implementations • 3 Aug 2023 • Otmane Sakhi, David Rohde, Nicolas Chopin
We compare our method to the commonly adopted Plackett-Luce policy class and demonstrate the effectiveness of our approach on problems with action space sizes in the order of millions.
no code implementations • 25 May 2023 • Imad Aouali, Victor-Emmanuel Brunel, David Rohde, Anna Korba
In particular, it is also valid for standard IPS without making the assumption that the importance weights are bounded.
1 code implementation • 5 Oct 2022 • Alexandre Gilotte, Ahmed Ben Yahmed, David Rohde
Aggregating a dataset, then injecting some noise, is a simple and common way to release differentially private data. However, aggregated data -- even without noise -- is not an appropriate input for machine learning classifiers. In this work, we show how a new model, similar to a logistic regression, may be learned from aggregated data only by approximating the unobserved feature distribution with a maximum entropy hypothesis.
no code implementations • 18 Sep 2022 • Imad Aouali, Amine Benhalloum, Martin Bompaire, Benjamin Heymann, Olivier Jeunen, David Rohde, Otmane Sakhi, Flavian vasile
Naturally, the reason for this is that we can directly measure utility metrics that rely on interventions, being the recommendations that are being shown to users.
no code implementations • 10 Aug 2022 • Imad Aouali, Achraf Ait Sidi Hammou, Sergey Ivanov, Otmane Sakhi, David Rohde, Flavian vasile
We introduce Probabilistic Rank and Reward (PRR), a scalable probabilistic model for personalized slate recommendation.
no code implementations • 8 Aug 2022 • Otmane Sakhi, David Rohde, Alexandre Gilotte
Personalised interactive systems such as recommender systems require selecting relevant items from massive catalogs dependent on context.
no code implementations • NeurIPS Workshop ICBINB 2021 • David Rohde
It is often argued that causal inference is a step that follows probabilistic estimation in a two step procedure, with a separate statistical estimation and causal inference step and each step is governed by its own principles.
no code implementations • 26 Jul 2021 • Imad Aouali, Sergey Ivanov, Mike Gartrell, David Rohde, Flavian vasile, Victor Zaytsev, Diego Legrand
In this paper, we formulate several Bayesian models that incorporate the reward signal (Reward model), the rank signal (Rank model), or both (Full model), for non-personalized slate recommendation.
no code implementations • 1 Sep 2020 • Philomène Chagniot, Flavian vasile, David Rohde
Recommender systems are often optimised for short-term reward: a recommendation is considered successful if a reward (e. g. a click) can be observed immediately after the recommendation.
no code implementations • 28 Aug 2020 • Otmane Sakhi, Stephen Bonner, David Rohde, Flavian vasile
In contrast, the organic signal is typically strong and covers most items, but is not always relevant to the recommendation task.
no code implementations • 2 Oct 2019 • Finnian Lattimore, David Rohde
The concept of causality has a controversial history.
no code implementations • 2 Oct 2019 • Otmane Sakhi, Stephen Bonner, David Rohde, Flavian vasile
The combination of the re-parameterization trick with the use of variational auto-encoders has caused a sensation in Bayesian deep learning, allowing the training of realistic generative models of images and has considerably increased our ability to use scalable latent variable models.
no code implementations • 18 Sep 2019 • Olivier Jeunen, Dmytro Mykhaylov, David Rohde, Flavian vasile, Alexandre Gilotte, Martin Bompaire
In order to handle this "bandit-feedback" setting, several Counterfactual Risk Minimisation (CRM) methods have been proposed in recent years, that attempt to estimate the performance of different policies on historical data.
no code implementations • 9 Sep 2019 • Nhan Nguyen-Thanh, Dana Marinca, Kinda Khawam, David Rohde, Flavian vasile, Elena Simona Lohan, Steven Martin, Dominique Quadri
The proposed method has been targeted to the problem of the product recommendation in the online advertising.
no code implementations • 26 Jul 2019 • Olivier Jeunen, David Rohde, Flavian vasile
The procedure aims to give an answer to the question: "Given the natural sequence of user-item interactions up to time t, can we predict which item the user will interact with at time t+1?".
no code implementations • 17 Jun 2019 • Finnian Lattimore, David Rohde
The concept of causality has a controversial history.
no code implementations • 17 Jun 2019 • David Rohde
It is common practice in using regression type models for inferring causal effects, that inferring the correct causal relationship requires extra covariates are included or ``adjusted for''.
no code implementations • pproximateinference AABI Symposium 2019 • David Rohde, Stephen Bonner
An attractive feature of the latent variable approach is that, as the user continues to act, the posterior on the user's state tightens reflecting the recommender system's increased knowledge about that user.
no code implementations • 24 Apr 2019 • Dmytro Mykhaylov, David Rohde, Flavian vasile, Martin Bompaire, Olivier Jeunen
There are three quite distinct ways to train a machine learning model on recommender system logs.
1 code implementation • 2 Aug 2018 • David Rohde, Stephen Bonner, Travis Dunlop, Flavian vasile, Alexandros Karatzoglou
Recommender Systems are becoming ubiquitous in many settings and take many forms, from product recommendation in e-commerce stores, to query suggestions in search engines, to friend recommendation in social networks.