End-to-End Session-Based Recommendation on GPU

Moreira, Gabriel de Souza Pereira; Rabhi, Sara; Ak, Ronay; Schifferer, Benedikt

doi:10.1145/3460231.3473322

Cited by 4 publications

(3 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, attention-based approaches [13,14,19,23,37] and graph neural network (GNN)-based methods [39][40][41][42]44] have been proposed to enhance the longer and deeper dependencies for SBRSs. Furthermore, Transformer [34]-based SBRSs [2,4,5,18] show the state-of-the-art prediction performance due to its powerful and efficient self-attention mechanism. Multi-task learning (MTL) [11,16,20] has been also adopted for SBRSs to enhance the next item prediction via generalization.…”

Section: Related Workmentioning

confidence: 99%

M2TRec: Metadata-aware Multi-task Transformer for Large-scale and Cold-start free Session-based Recommendations

Shalaby

Afsharinejad

et al. 2022

Proceedings of the 16th ACM Conference on Recommender Systems

View full text Add to dashboard Cite

Session-based recommender systems (SBRSs) have shown superior performance over conventional methods. However, they show limited scalability on large-scale industrial datasets since most models learn one embedding per item. This leads to a large memory requirement (of storing one vector per item) and poor performance on sparse sessions with cold-start or unpopular items. Using one public and one large industrial dataset, we experimentally show that state-of-the-art SBRSs have low performance on sparse sessions with sparse items. We propose M2TRec, a Metadata-aware Multitask Transformer model for session-based recommendations. Our proposed method learns a transformation function from item metadata to embeddings, and is thus, item-ID free (i.e., does not need to learn one embedding per item). It integrates item metadata to learn shared representations of diverse item attributes. During inference, new or unpopular items will be assigned identical representations for the attributes they share with items previously observed during training, and thus will have similar representations with those items, enabling recommendations of even cold-start and sparse items. Additionally, M2TRec is trained in a multi-task setting to predict the next item in the session along with its primary category and subcategories. Our multi-task strategy makes the model converge faster and significantly improves the overall performance. Experimental results show significant performance gains using our proposed approach on sparse items on the two datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

M2TRec: Metadata-aware Multi-task Transformer for Large-scale and Cold-start free Session-based Recommendations

Shalaby

Afsharinejad

et al. 2022

Proceedings of the 16th ACM Conference on Recommender Systems

View full text Add to dashboard Cite

show abstract

“…Regarding SBRSs, to the best of our knowledge, there is only one comprehensive survey on SBRSs [12] to systematically discuss the session-based recommendation problem, data characteristics, recent progress, approach taxonomy, applications and future directions. Ludewig et al [5] conducted an empirical study on some representative SBRS algorithms while Gabriel et al [1] provided a tutorial on session-based recommendation on GPU.…”

Section: Related Workmentioning

confidence: 99%

“…There is a lack of a unified framework to well categorize them, and there are no unified problem statements for the research problem(s) [12]. A few tutorials have focused on sequence-aware recommender systems [9], deep learning-based sequential recommendations [2], and session-based recommendation on GPU [1]. However, there is no work to provide a unified framework and problem statement to remove the commonly existing and various inconsistencies in the areas of SRSs and SBRSs.…”

Section: Introductionmentioning

confidence: 99%

Sequential/Session-based Recommendations: Challenges, Approaches, Applications and Opportunities

Wang,

Zhang,

et al. 2022

Preprint

View full text Add to dashboard Cite

In recent years, sequential recommender systems (SRSs) and sessionbased recommender systems (SBRSs) have emerged as a new paradigm of RSs to capture users' short-term but dynamic preferences for enabling more timely and accurate recommendations. Although SRSs and SBRSs have been extensively studied, there are many inconsistencies in this area caused by the diverse descriptions, settings, assumptions and application domains. There is no work to provide a unified framework and problem statement to remove the commonly existing and various inconsistencies in the area of SR/SBR. There is a lack of work to provide a comprehensive and systematic demonstration of the data characteristics, key challenges, most representative and state-of-the-art approaches, typical realworld applications and important future research directions in the area. This work aims to fill in these gaps so as to facilitate further research in this exciting and vibrant area. CCS CONCEPTS• Information systems → Recommender systems.

show abstract

Sequential/Session-based Recommendations

Wang

Zhang²,

et al. 2022

Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

End-to-End Session-Based Recommendation on GPU

Cited by 4 publications

References 2 publications

M2TRec: Metadata-aware Multi-task Transformer for Large-scale and Cold-start free Session-based Recommendations

M2TRec: Metadata-aware Multi-task Transformer for Large-scale and Cold-start free Session-based Recommendations

Sequential/Session-based Recommendations: Challenges, Approaches, Applications and Opportunities

Sequential/Session-based Recommendations

Contact Info

Product

Resources

About