Ing. Petr Kasalický

Článek

10.1109/TNNLS.2025.3597051

Katedra aplikované matematiky

We introduce a new convolutional autoencoder architecture for user modeling and recommendation tasks with several improvements over the state of the art. First, our model has the flexibility to learn a set of associations and combinations between different interaction types in a way that carries over to each user and item. Second, our model is able to learn jointly from both the explicit ratings and the implicit information in the sampling pattern (which we refer to as ”implicit feedback”). It can also make separate predictions for the probability of consuming content and the likelihood of granting it a high rating if observed. This not only allows the model to make predictions for both the implicit and explicit feedback, but also increases the informativeness of the predictions: in particular, our model can identify items that users would not have been likely to consume naturally, but would be likely to enjoy if exposed to them. Finally, we provide several generalization bounds for our model, which, to the best of our knowledge, are among the first generalization bounds for autoencoders in a Recommender systems setting; we also show that optimizing our loss function guarantees the recovery of the exact sampling distribution over interactions up to a small error in total variation. In experiments on several real-life datasets, we achieve state-of-the-art performance on both the implicit and explicit feedback prediction tasks despite relying on a single model for both, and benefiting from additional interpretability in the form of individual predictions for the probabilities of each possible rating.

Stať ve sborníku

10.1145/3705328.3748147

Katedra aplikované matematiky

Industry-scale recommender systems face a core challenge: representing entities with high cardinality, such as users or items, using dense embeddings that must be accessible during both training and inference. However, as embedding sizes grow, memory constraints make storage and access increasingly difficult. We describe a lightweight, learnable embedding compression technique that projects dense embeddings into a high-dimensional, sparsely activated space. Designed for retrieval tasks, our method reduces memory requirements while preserving retrieval performance, enabling scalable deployment under strict resource constraints. Our results demonstrate that leveraging sparsity is a promising approach for improving the efficiency of large-scale recommenders. We release our code at https://github.com/recombee/CompresSAE.

Stať ve sborníku

Katedra aplikované matematiky

The evaluation of recommendation systems is a complex task. The offline and online evaluation metrics for recommender systems are ambiguous in their true objectives. The majority of recently published papers benchmark their methods using ill-posed offline evaluation methodology that often fails to predict true online performance. Because of this, the impact that academic research has on the industry is reduced. The aim of our research is to investigate and compare the online performance of offline evaluation metrics. We show that penalizing popular items and considering the time of transactions during the evaluation significantly improves our ability to choose the best recommendation model for a live recommender system. Our results, averaged over five large-size real-world live data procured from recommenders, aim to help the academic community to understand better offline evaluation and optimization criteria that are more relevant for real applications of recommender systems.

Stať ve sborníku

10.1145/3604915.3610654

Katedra aplikované matematiky

We propose a robust recommender systems model which performs matrix completion and a ratings-wise uncertainty estimation jointly. Whilst the prediction module is purely based on an implicit low-rank assumption imposed via nuclear norm regularization, our loss function is augmented by an uncertainty estimation module which learns an anomaly score for each individual rating via a Graph Neural Network: data points deemed more anomalous by the GNN are downregulated in the loss function used to train the low-rank module. The whole model is trained in an end-to-end fashion, allowing the anomaly detection module to tap on the supervised information available in the form of ratings. Thus, our model’s predictors enjoy the favourable generalization properties that come with being chosen from small function space (i.e., low-rank matrices), whilst exhibiting the robustness to outliers and flexibility that comes with deep learning methods. Furthermore, the anomaly scores themselves contain valuable qualitative information. Experiments on various real-life datasets demonstrate that our model outperforms standard matrix completion and other baselines, confirming the usefulness of the anomaly detection module.

Stať ve sborníku

10.1145/3523227.3551482

Katedra aplikované matematiky

Recently, the RS research community has witnessed a surge in popularity for shallow autoencoder-based CF methods. Due to its straightforward implementation and high accuracy on item retrieval metrics, EASE is potentially the most prominent of these models. Despite its accuracy and simplicity, EASE cannot be employed in some real-world recommender system applications due to its inability to scale to huge interaction matrices. In this paper, we proposed ELSA, a scalable shallow autoencoder method for implicit feedback recommenders. ELSA is a scalable autoencoder in which the hidden layer is factorizable into a low-rank plus sparse structure, thereby drastically lowering memory consumption and computation time. We conducted a comprehensive offline experimental section that combined synthetic and several real-world datasets. We also validated our strategy in an online setting by comparing ELSA to baselines in a live recommender system using an A/B test. Experiments demonstrate that ELSA is scalable and has competitive performance. Finally, we demonstrate the explainability of ELSA by illustrating the recovered latent space.