no code implementations • JEP/TALN/RECITAL 2021 • Elena V. Epure, Guillaume Salha-Galvan, Manuel Moussallam, Romain Hennequin
Nous résumons nos travaux de recherche, présentés à la conférence EMNLP 2020 et portant sur la modélisation de la perception des genres musicaux à travers différentes cultures, à partir de représentations sémantiques spécifiques à différentes langues.
1 code implementation • 5 Jan 2024 • Léa Briand, Théo Bontempelli, Walid Bendada, Mathieu Morlon, François Rigaud, Benjamin Chapus, Thomas Bouabça, Guillaume Salha-Galvan
This paper presents our recent initiatives to foster the discoverability of new releases on the music streaming service Deezer.
1 code implementation • 24 Aug 2023 • Walid Bendada, Guillaume Salha-Galvan, Romain Hennequin, Thomas Bouabça, Tristan Cazenave
A prevalent practice in recommender systems consists in averaging item embeddings to represent users or higher-level concepts in the same embedding space.
no code implementations • 6 Jul 2023 • Walid Bendada, Théo Bontempelli, Mathieu Morlon, Benjamin Chapus, Thibault Cador, Thomas Bouabça, Guillaume Salha-Galvan
This paper introduces Track Mix, a personalized playlist generation system released in 2022 on the music streaming service Deezer.
1 code implementation • 17 Apr 2023 • Viet-Anh Tran, Guillaume Salha-Galvan, Bruno Sguerra, Romain Hennequin
Transformers emerged as powerful methods for sequential recommendation.
1 code implementation • 12 Apr 2023 • Walid Bendada, Guillaume Salha-Galvan, Thomas Bouabça, Tristan Cazenave
Music streaming services often aim to recommend songs for users to extend the playlists they have created on these services.
1 code implementation • 16 Nov 2022 • Guillaume Salha-Galvan, Johannes F. Lutzeyer, George Dasoulas, Romain Hennequin, Michalis Vazirgiannis
It is still unclear to what extent one can improve CD with GAE and VGAE, especially in the absence of node features.
no code implementations • 15 Jul 2022 • Théo Bontempelli, Benjamin Chapus, François Rigaud, Mathieu Morlon, Marin Lorant, Guillaume Salha-Galvan
We detail the motivations, the development, and the deployment of this system on Deezer.
no code implementations • 29 May 2022 • Guillaume Salha-Galvan
In the last part of this thesis, we evaluate our methods on several graphs extracted from the music streaming service Deezer.
1 code implementation • 2 Feb 2022 • Guillaume Salha-Galvan, Johannes F. Lutzeyer, George Dasoulas, Romain Hennequin, Michalis Vazirgiannis
Graph autoencoders (GAE) and variational graph autoencoders (VGAE) emerged as powerful methods for link prediction.
1 code implementation • 2 Aug 2021 • Guillaume Salha-Galvan, Romain Hennequin, Benjamin Chapus, Viet-Anh Tran, Michalis Vazirgiannis
In this paper, we model this cold start similar artists ranking problem as a link prediction task in a directed and attributed graph, connecting artists to their top-k most similar neighbors and incorporating side musical information.
1 code implementation • 26 Jul 2021 • Viet-Anh Tran, Guillaume Salha-Galvan, Romain Hennequin, Manuel Moussallam
Existing extensions of CML also either ignore the heterogeneity of user-item relations, i. e. that a user can simultaneously like very different items, or the latent item-item relations, i. e. that a user's preference for an item depends, not only on its intrinsic characteristics, but also on items they previously interacted with.
1 code implementation • 7 Jun 2021 • Léa Briand, Guillaume Salha-Galvan, Walid Bendada, Mathieu Morlon, Viet-Anh Tran
This is commonly referred to as the user cold start problem.