Alert button
Picture for Guillaume Salha-Galvan

Guillaume Salha-Galvan

Alert button

On the Consistency of Average Embeddings for Item Recommendation

Aug 30, 2023
Walid Bendada, Guillaume Salha-Galvan, Romain Hennequin, Thomas Bouabça, Tristan Cazenave

Figure 1 for On the Consistency of Average Embeddings for Item Recommendation
Figure 2 for On the Consistency of Average Embeddings for Item Recommendation
Figure 3 for On the Consistency of Average Embeddings for Item Recommendation
Figure 4 for On the Consistency of Average Embeddings for Item Recommendation

A prevalent practice in recommender systems consists in averaging item embeddings to represent users or higher-level concepts in the same embedding space. This paper investigates the relevance of such a practice. For this purpose, we propose an expected precision score, designed to measure the consistency of an average embedding relative to the items used for its construction. We subsequently analyze the mathematical expression of this score in a theoretical setting with specific assumptions, as well as its empirical behavior on real-world data from music streaming services. Our results emphasize that real-world averages are less consistent for recommendation, which paves the way for future research to better align real-world embeddings with assumptions from our theoretical setting.

* 17th ACM Conference on Recommender Systems (RecSys 2023) 
Viaarxiv icon

Track Mix Generation on Music Streaming Services using Transformers

Jul 06, 2023
Walid Bendada, Théo Bontempelli, Mathieu Morlon, Benjamin Chapus, Thibault Cador, Thomas Bouabça, Guillaume Salha-Galvan

Figure 1 for Track Mix Generation on Music Streaming Services using Transformers
Figure 2 for Track Mix Generation on Music Streaming Services using Transformers
Figure 3 for Track Mix Generation on Music Streaming Services using Transformers

This paper introduces Track Mix, a personalized playlist generation system released in 2022 on the music streaming service Deezer. Track Mix automatically generates "mix" playlists inspired by initial music tracks, allowing users to discover music similar to their favorite content. To generate these mixes, we consider a Transformer model trained on millions of track sequences from user playlists. In light of the growing popularity of Transformers in recent years, we analyze the advantages, drawbacks, and technical challenges of using such a model for mix generation on the service, compared to a more traditional collaborative filtering approach. Since its release, Track Mix has been generating playlists for millions of users daily, enhancing their music discovery experience on Deezer.

* RecSys 2023 - Industry track with oral presentation 
Viaarxiv icon

Attention Mixtures for Time-Aware Sequential Recommendation

Apr 17, 2023
Viet-Anh Tran, Guillaume Salha-Galvan, Bruno Sguerra, Romain Hennequin

Figure 1 for Attention Mixtures for Time-Aware Sequential Recommendation
Figure 2 for Attention Mixtures for Time-Aware Sequential Recommendation

Transformers emerged as powerful methods for sequential recommendation. However, existing architectures often overlook the complex dependencies between user preferences and the temporal context. In this short paper, we introduce MOJITO, an improved Transformer sequential recommender system that addresses this limitation. MOJITO leverages Gaussian mixtures of attention-based temporal context and item embedding representations for sequential modeling. Such an approach permits to accurately predict which items should be recommended next to users depending on past actions and the temporal context. We demonstrate the relevance of our approach, by empirically outperforming existing Transformers for sequential recommendation on several real-world datasets.

* SIGIR 2023 
Viaarxiv icon

A Scalable Framework for Automatic Playlist Continuation on Music Streaming Services

Apr 12, 2023
Walid Bendada, Guillaume Salha-Galvan, Thomas Bouabça, Tristan Cazenave

Figure 1 for A Scalable Framework for Automatic Playlist Continuation on Music Streaming Services
Figure 2 for A Scalable Framework for Automatic Playlist Continuation on Music Streaming Services
Figure 3 for A Scalable Framework for Automatic Playlist Continuation on Music Streaming Services
Figure 4 for A Scalable Framework for Automatic Playlist Continuation on Music Streaming Services

Music streaming services often aim to recommend songs for users to extend the playlists they have created on these services. However, extending playlists while preserving their musical characteristics and matching user preferences remains a challenging task, commonly referred to as Automatic Playlist Continuation (APC). Besides, while these services often need to select the best songs to recommend in real-time and among large catalogs with millions of candidates, recent research on APC mainly focused on models with few scalability guarantees and evaluated on relatively small datasets. In this paper, we introduce a general framework to build scalable yet effective APC models for large-scale applications. Based on a represent-then-aggregate strategy, it ensures scalability by design while remaining flexible enough to incorporate a wide range of representation learning and sequence modeling techniques, e.g., based on Transformers. We demonstrate the relevance of this framework through in-depth experimental validation on Spotify's Million Playlist Dataset (MPD), the largest public dataset for APC. We also describe how, in 2022, we successfully leveraged this framework to improve APC in production on Deezer. We report results from a large-scale online A/B test on this service, emphasizing the practical impact of our approach in such a real-world application.

* Accepted as a Full Paper at the SIGIR 2023 conference 
Viaarxiv icon

New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction

Nov 16, 2022
Guillaume Salha-Galvan, Johannes F. Lutzeyer, George Dasoulas, Romain Hennequin, Michalis Vazirgiannis

Figure 1 for New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction
Figure 2 for New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction
Figure 3 for New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction
Figure 4 for New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction

Graph autoencoders (GAE) and variational graph autoencoders (VGAE) emerged as powerful methods for link prediction (LP). Their performances are less impressive on community detection (CD), where they are often outperformed by simpler alternatives such as the Louvain method. It is still unclear to what extent one can improve CD with GAE and VGAE, especially in the absence of node features. It is moreover uncertain whether one could do so while simultaneously preserving good performances on LP in a multi-task setting. In this workshop paper, summarizing results from our journal publication (Salha-Galvan et al. 2022), we show that jointly addressing these two tasks with high accuracy is possible. For this purpose, we introduce a community-preserving message passing scheme, doping our GAE and VGAE encoders by considering both the initial graph and Louvain-based prior communities when computing embedding spaces. Inspired by modularity-based clustering, we further propose novel training and optimization strategies specifically designed for joint LP and CD. We demonstrate the empirical effectiveness of our approach, referred to as Modularity-Aware GAE and VGAE, on various real-world graphs.

* This NeurIPS 2022 GLFrontiers workshop paper summarizes results from the following journal article: arXiv:2202.00961. arXiv admin note: text overlap with arXiv:2205.14651 
Viaarxiv icon

Flow Moods: Recommending Music by Moods on Deezer

Jul 15, 2022
Théo Bontempelli, Benjamin Chapus, François Rigaud, Mathieu Morlon, Marin Lorant, Guillaume Salha-Galvan

Figure 1 for Flow Moods: Recommending Music by Moods on Deezer
Figure 2 for Flow Moods: Recommending Music by Moods on Deezer

The music streaming service Deezer extensively relies on its Flow algorithm, which generates personalized radio-style playlists of songs, to help users discover musical content. Nonetheless, despite promising results over the past years, Flow used to ignore the moods of users when providing recommendations. In this paper, we present Flow Moods, an improved version of Flow that addresses this limitation. Flow Moods leverages collaborative filtering, audio content analysis, and mood annotations from professional music curators to generate personalized mood-specific playlists at scale. We detail the motivations, the development, and the deployment of this system on Deezer. Since its release in 2021, Flow Moods has been recommending music by moods to millions of users every day.

* 16th ACM Conference on Recommender Systems (RecSys 2022) - Industry paper 
Viaarxiv icon

Contributions to Representation Learning with Graph Autoencoders and Applications to Music Recommendation

Jun 06, 2022
Guillaume Salha-Galvan

Graph autoencoders (GAE) and variational graph autoencoders (VGAE) emerged as two powerful groups of unsupervised node embedding methods, with various applications to graph-based machine learning problems such as link prediction and community detection. Nonetheless, at the beginning of this Ph.D. project, GAE and VGAE models were also suffering from key limitations, preventing them from being adopted in the industry. In this thesis, we present several contributions to improve these models, with the general aim of facilitating their use to address industrial-level problems involving graph representations. Firstly, we propose two strategies to overcome the scalability issues of previous GAE and VGAE models, permitting to effectively train these models on large graphs with millions of nodes and edges. These strategies leverage graph degeneracy and stochastic subgraph decoding techniques, respectively. Besides, we introduce Gravity-Inspired GAE and VGAE, providing the first extensions of these models for directed graphs, that are ubiquitous in industrial applications. We also consider extensions of GAE and VGAE models for dynamic graphs. Furthermore, we argue that GAE and VGAE models are often unnecessarily complex, and we propose to simplify them by leveraging linear encoders. Lastly, we introduce Modularity-Aware GAE and VGAE to improve community detection on graphs, while jointly preserving good performances on link prediction. In the last part of this thesis, we evaluate our methods on several graphs extracted from the music streaming service Deezer. We put the emphasis on graph-based music recommendation problems. In particular, we show that our methods can improve the detection of communities of similar musical items to recommend to users, that they can effectively rank similar artists in a cold start setting, and that they permit modeling the music genre perception across cultures.

* Ph.D. thesis defended at \'Ecole Polytechnique (IPP) in March 2022. As mentioned in this thesis, several chapters present results also published in scientific articles written with co-authors 
Viaarxiv icon

Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction

Feb 02, 2022
Guillaume Salha-Galvan, Johannes F. Lutzeyer, George Dasoulas, Romain Hennequin, Michalis Vazirgiannis

Figure 1 for Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction
Figure 2 for Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction
Figure 3 for Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction
Figure 4 for Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction

Graph autoencoders (GAE) and variational graph autoencoders (VGAE) emerged as powerful methods for link prediction. Their performances are less impressive on community detection problems where, according to recent and concurring experimental evaluations, they are often outperformed by simpler alternatives such as the Louvain method. It is currently still unclear to which extent one can improve community detection with GAE and VGAE, especially in the absence of node features. It is moreover uncertain whether one could do so while simultaneously preserving good performances on link prediction. In this paper, we show that jointly addressing these two tasks with high accuracy is possible. For this purpose, we introduce and theoretically study a community-preserving message passing scheme, doping our GAE and VGAE encoders by considering both the initial graph structure and modularity-based prior communities when computing embedding spaces. We also propose novel training and optimization strategies, including the introduction of a modularity-inspired regularizer complementing the existing reconstruction losses for joint link prediction and community detection. We demonstrate the empirical effectiveness of our approach, referred to as Modularity-Aware GAE and VGAE, through in-depth experimental validation on various real-world graphs.

* Under review 
Viaarxiv icon