Pessimistic reward models for off-policy learning in recommendation
O Jeunen, B Goethals - Proceedings of the 15th ACM Conference on …, 2021 - dl.acm.org
Methods for bandit learning from user interactions often require a model of the reward a
certain context-action pair will yield–for example, the probability of a click on a …
certain context-action pair will yield–for example, the probability of a click on a …
Pessimistic decision-making for recommender systems
O Jeunen, B Goethals - ACM Transactions on Recommender Systems, 2023 - dl.acm.org
Modern recommender systems are often modelled under the sequential decision-making
paradigm, where the system decides which recommendations to show in order to maximise …
paradigm, where the system decides which recommendations to show in order to maximise …
Top-k contextual bandits with equity of exposure
O Jeunen, B Goethals - Proceedings of the 15th ACM Conference on …, 2021 - dl.acm.org
The contextual bandit paradigm provides a general framework for decision-making under
uncertainty. It is theoretically well-defined and well-studied, and many personalisation use …
uncertainty. It is theoretically well-defined and well-studied, and many personalisation use …
Joint policy-value learning for recommendation
Conventional approaches to recommendation often do not explicitly take into account
information on previously shown recommendations and their recorded responses. One …
information on previously shown recommendations and their recorded responses. One …
On (Normalised) Discounted Cumulative Gain as an Off-Policy Evaluation Metric for Top-n Recommendation
O Jeunen, I Potapov, A Ustimenko - Proceedings of the 30th ACM …, 2024 - dl.acm.org
Approaches to recommendation are typically evaluated in one of two ways:(1) via a
(simulated) online experiment, often seen as the gold standard, or (2) via some offline …
(simulated) online experiment, often seen as the gold standard, or (2) via some offline …
How Important is Periodic Model update in Recommender System?
In real-world recommender model deployments, the models are typically retrained and
deployed repeatedly. It is the rule-of-thumb to periodically retrain recommender models to …
deployed repeatedly. It is the rule-of-thumb to periodically retrain recommender models to …
MARS-Gym: A Gym framework to model, train, and evaluate Recommender Systems for Marketplaces
MRO Santana, LC Melo, FHF Camargo… - … Conference on Data …, 2020 - ieeexplore.ieee.org
Recommender Systems are especially challenging for marketplaces since they must
maximize user satisfaction while maintaining the healthiness and fairness of such …
maximize user satisfaction while maintaining the healthiness and fairness of such …
Offline approaches to recommendation with online success
O Jeunen - 2021 - repository.uantwerpen.be
Recommender systems are information retrieval applications that provide users with
algorithmic recommendations, in order to assist decision-making when sufficient knowledge …
algorithmic recommendations, in order to assist decision-making when sufficient knowledge …
User-centric evaluation of recommender systems: a literature review
K Nanath, M Ahmed - International Journal of Business …, 2023 - inderscienceonline.com
Recommender systems have seen a rapid rise of application in various industries, with
several services now being implemented online. Over the years, various authors have been …
several services now being implemented online. Over the years, various authors have been …
A gentle introduction to recommendation as counterfactual policy learning
The objective of this tutorial is to give a structured overview of the conceptual frameworks
behind current state-of-the-art recommender systems, explain their underlying assumptions …
behind current state-of-the-art recommender systems, explain their underlying assumptions …