1 code implementation • 22 Jan 2023 • Khalil Damak, Sami Khenissi, Olfa Nasraoui
The most common approach to mitigating exposure bias in recommendation has been Inverse Propensity Scoring (IPS), which consists of down-weighting the interacted predictions in the loss function in proportion to their propensities of exposure, yielding a theoretically unbiased learning.
1 code implementation • 30 Jul 2021 • Khalil Damak, Sami Khenissi, Olfa Nasraoui
In this work, we first propose a novel explainable loss function and a corresponding Matrix Factorization-based model called Explainable Bayesian Personalized Ranking (EBPR) that generates recommendations along with item-based explanations.
1 code implementation • 21 Aug 2020 • Sami Khenissi, Mariem Boujelbene, Olfa Nasraoui
We present a theoretical framework to model the asymptotic evolution of the different components of a recommender system operating within a feedback loop setting, and derive theoretical bounds and convergence properties on quantifiable measures of the user discovery and blind spots.
no code implementations • 1 Jan 2020 • Sami Khenissi, Olfa Nasraoui
Then we model the exposure that is borne from the interaction between the user and the recommender system and propose new debiasing strategies for these systems.