On the Mechanisms of Collaborative Learning in VAE Recommenders

• URL: http://arxiv.org/abs/2511.06781v1
• Date: Mon, 10 Nov 2025 07:13:58 GMT
• Title: On the Mechanisms of Collaborative Learning in VAE Recommenders
• Authors: Tung-Long Vuong, Julien Monteil, Hien Dang, Volodymyr Vaskovych, Trung Le, Vu Nguyen,
• Abstract summary: Variational Autoencoders (VAEs) are a powerful alternative to matrix factorization for recommendation.<n>We analyze how collaboration arises in VAE-based collaborative filtering (CF) and show it is governed by latent proximity.<n>Our analyses are validated on the Netflix, MovieLens-20M, and Million Song datasets.
• Score: 12.77626224639472
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Variational Autoencoders (VAEs) are a powerful alternative to matrix factorization for recommendation. A common technique in VAE-based collaborative filtering (CF) consists in applying binary input masking to user interaction vectors, which improves performance but remains underexplored theoretically. In this work, we analyze how collaboration arises in VAE-based CF and show it is governed by latent proximity: we derive a latent sharing radius that informs when an SGD update on one user strictly reduces the loss on another user, with influence decaying as the latent Wasserstein distance increases. We further study the induced geometry: with clean inputs, VAE-based CF primarily exploits \emph{local} collaboration between input-similar users and under-utilizes global collaboration between far-but-related users. We compare two mechanisms that encourage \emph{global} mixing and characterize their trade-offs: (1) $\beta$-KL regularization directly tightens the information bottleneck, promoting posterior overlap but risking representational collapse if too large; (2) input masking induces stochastic geometric contractions and expansions, which can bring distant users onto the same latent neighborhood but also introduce neighborhood drift. To preserve user identity while enabling global consistency, we propose an anchor regularizer that aligns user posteriors with item embeddings, stabilizing users under masking and facilitating signal sharing across related items. Our analyses are validated on the Netflix, MovieLens-20M, and Million Song datasets. We also successfully deployed our proposed algorithm on an Amazon streaming platform following a successful online experiment.

Related papers

• Leave No One Behind: Fairness-Aware Cross-Domain Recommender Systems for Non-Overlapping Users [13.420661387194148]
  Cross-domain recommendation (CDR) methods leverage overlapping users to transfer knowledge from a source domain to a target domain.<n>We propose a novel solution that generates virtual source-domain users for non-overlapping target-domain users.<n>Our method effectively mitigates the CDR non-overlapping user bias, without loss of overall accuracy.
  arXiv  Detail & Related papers  (2025-07-23T17:59:08Z)
• Online Clustering of Dueling Bandits [59.09590979404303]
  We introduce the first "clustering of dueling bandit algorithms" to enable collaborative decision-making based on preference feedback.<n>We propose two novel algorithms: (1) Clustering of Linear Dueling Bandits (COLDB) which models the user reward functions as linear functions of the context vectors, and (2) Clustering of Neural Dueling Bandits (CONDB) which uses a neural network to model complex, non-linear user reward functions.
  arXiv  Detail & Related papers  (2025-02-04T07:55:41Z)
• EDGE-Rec: Efficient and Data-Guided Edge Diffusion For Recommender Systems Graphs [0.0]
  We propose a new attention mechanism to take advantage of real-valued interaction weights as well as user and item features directly. We train a novel Graph Diffusion Transformer GDiT architecture to iteratively denoise the weighted interaction matrix of the user-item interaction graph directly. Inspired by the recent progress in text-conditioned image generation, our method directly produces user-item rating predictions on the same scale as the original ratings.
  arXiv  Detail & Related papers  (2024-09-23T03:23:20Z)
• Personalized Federated Collaborative Filtering: A Variational AutoEncoder Approach [49.63614966954833]
  Federated Collaborative Filtering (FedCF) is an emerging field focused on developing a new recommendation framework with preserving privacy.<n>Existing FedCF methods typically combine distributed Collaborative Filtering (CF) algorithms with privacy-preserving mechanisms, and then preserve personalized information into a user embedding vector.<n>This paper proposes a novel personalized FedCF method by preserving users' personalized information into a latent variable and a neural model simultaneously.
  arXiv  Detail & Related papers  (2024-08-16T05:49:14Z)
• Graph Collaborative Signals Denoising and Augmentation for Recommendation [75.25320844036574]
  We propose a new graph adjacency matrix that incorporates user-user and item-item correlations. We show that the inclusion of user-user and item-item correlations can improve recommendations for users with both abundant and insufficient interactions.
  arXiv  Detail & Related papers  (2023-04-06T19:43:37Z)
• ByzSecAgg: A Byzantine-Resistant Secure Aggregation Scheme for Federated Learning Based on Coded Computing and Vector Commitment [61.540831911168226]
  ByzSecAgg is an efficient secure aggregation scheme for federated learning.<n>ByzSecAgg is resistant to Byzantine attacks and privacy leakages.
  arXiv  Detail & Related papers  (2023-02-20T11:15:18Z)
• Bootstrapping User and Item Representations for One-Class Collaborative Filtering [24.30834981766022]
  One-class collaborative filtering (OCCF) is to identify user-item pairs that are positively-related but have not been interacted yet. This paper proposes a novel OCCF framework, named as BUIR, which does not require negative sampling.
  arXiv  Detail & Related papers  (2021-05-13T14:24:13Z)
• Modular Interactive Video Object Segmentation: Interaction-to-Mask, Propagation and Difference-Aware Fusion [68.45737688496654]
  We present a modular interactive VOS framework which decouples interaction-to-mask and mask propagation. We show that our method outperforms current state-of-the-art algorithms while requiring fewer frame interactions.
  arXiv  Detail & Related papers  (2021-03-14T14:39:08Z)
• Robust Person Re-Identification through Contextual Mutual Boosting [77.1976737965566]
  We propose the Contextual Mutual Boosting Network (CMBN) to localize pedestrians. It localizes pedestrians and recalibrates features by effectively exploiting contextual information and statistical inference. Experiments on the benchmarks demonstrate the superiority of the architecture compared the state-of-the-art.
  arXiv  Detail & Related papers  (2020-09-16T06:33:35Z)
• Leveraging Cross Feedback of User and Item Embeddings with Attention for Variational Autoencoder based Collaborative Filtering [25.569930256022925]
  We propose a VAE-based Bayesian MF framework to approximate the user-item joint distribution. The framework is iterative with cross feedback of user and item embeddings into each other's encoders. It reconstructs the data via the matrix factorization over the currently re-sampled user and item embeddings.
  arXiv  Detail & Related papers  (2020-02-21T06:05:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.