Attentive Item2vec: Neural Attentive User Representations

Fourteenth ACM Conference on Recommender Systems

Fuchs

Caciularu

et al. 2020

Self Cite

Two main challenges in recommender systems are modeling users with heterogeneous taste, and providing explainable recommendations. In this paper, we propose the neural Attentive Multi-Persona Collaborative Filtering (AMP-CF) model as a unified solution for both problems. AMP-CF breaks down the user to several latent 'personas' (profiles) that identify and discern the different tastes and inclinations of the user. Then, the revealed personas are used to generate and explain the final recommendation list for the user. AMP-CF models users as an attentive mixture of personas, enabling a dynamic user representation that changes based on the item under consideration. We demonstrate AMP-CF on five collaborative filtering datasets from the domains of movies, music, video games and social networks. As an additional contribution, we propose a novel evaluation scheme for comparing the different items in a recommendation list based on the distance from the underlying distribution of "tastes" in the user's historical items. Experimental results show that AMP-CF is competitive with other state-of-the-art models. Finally, we provide qualitative results to showcase the ability of AMP-CF to explain its recommendations. CCS Concepts: • Computing methodologies → Learning latent representations; • Information systems → Collaborative filtering.

Section: Related Workmentioning

confidence: 99%

“…Items are clustered together in prioritymedoids and cover-trees are used to declaratively capture a desired balance between predicted rating and diversity. In [3], DiRec was compared to other algorithms such as Greedy and Swap [41] and showed to significantly outperformed them on several evaluation metrics.…”

Section: Evaluated Modelsmentioning

confidence: 99%

Explainable Recommendations via Attentive Multi-Persona Collaborative Filtering

Fourteenth ACM Conference on Recommender Systems

Fuchs

Caciularu

et al. 2020

Self Cite

“…Transformer-based models have revolutionized the fields of natural language processing [22,33,46] and recommender systems [34,44], improving upon other neural embedding models [3, 5, 7, 8, 13-15, 17, 36, 37, 40]. Significant strides were made in tasks such as machine translation [46], sentiment analysis [49], semantic textual similarity [16,23,35], and item similarity [6,11,12,28,34]. However, transformers employ a complex attention-based architecture comprising hundreds of millions of parameters that cannot be decomposed into smaller more interpretable components.…”

Section: Introductionmentioning

confidence: 99%

Interpreting BERT-based Text Similarity via Activation and Saliency Maps

Malkiel

Ginzburg

Proceedings of the ACM Web Conference 2022

et al. 2022

Self Cite

Recently, there has been growing interest in the ability of Transformerbased models to produce meaningful embeddings of text with several applications, such as text similarity. Despite significant progress in the field, the explanations for similarity predictions remain challenging, especially in unsupervised settings. In this work, we present an unsupervised technique for explaining paragraph similarities inferred by pre-trained BERT models. By looking at a pair of paragraphs, our technique identifies important words that dictate each paragraph's semantics, matches between the words in both paragraphs, and retrieves the most important pairs that explain the similarity between the two. The method, which has been assessed by extensive human evaluations and demonstrated on datasets comprising long and complex paragraphs, has shown great promise, providing accurate interpretations that correlate better with human perceptions.

“…Neural attention mechanisms are emerging techniques that affects a variety of different fields [2], [11], [19], [21], [31], [32], [33]. For instance, in [24] the authors suggest using attention mechanism for improving the nearest neighbors in each view, however, the integration between different views is set by hyperparameters.…”

Section: Introduction and Related Workmentioning

confidence: 99%

Neural Attentive Multiview Machines

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Katz

Koenigstein

2020

Self Cite

An important problem in multiview representation learning is finding the optimal combination of views with respect to the specific task at hand. To this end, we introduce NAM: a Neural Attentive Multiview machine that learns multiview item representations and similarity by employing a novel attention mechanism. NAM harnesses multiple information sources and automatically quantifies their relevancy with respect to a supervised task. Finally, a very practical advantage of NAM is its robustness to the case of dataset with missing views. We demonstrate the effectiveness of NAM for the task of movies and app recommendations. Our evaluations indicate that NAM outperforms single view models as well as alternative multiview methods on item recommendations tasks, including cold-start scenarios.Index Terms -neural multiview attention machines, neural attention mechanisms, multiview representation learning, collaborative filtering, recommender systems, item similarity