Optimizing Sequential Recommendation Models with Scaling Laws and Approximate Entropy

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Shen, Tingjia, Wang, Hao, Wu, Chuhan, Chin, Jin Yao, Guo, Wei, Liu, Yong, Guo, Huifeng, Lian, Defu, Tang, Ruiming, Chen, Enhong
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912237197721600
author Shen, Tingjia
Wang, Hao
Wu, Chuhan
Chin, Jin Yao
Guo, Wei
Liu, Yong
Guo, Huifeng
Lian, Defu
Tang, Ruiming
Chen, Enhong
author_facet Shen, Tingjia
Wang, Hao
Wu, Chuhan
Chin, Jin Yao
Guo, Wei
Liu, Yong
Guo, Huifeng
Lian, Defu
Tang, Ruiming
Chen, Enhong
contents Scaling Laws have emerged as a powerful framework for understanding how model performance evolves as they increase in size, providing valuable insights for optimizing computational resources. In the realm of Sequential Recommendation (SR), which is pivotal for predicting users' sequential preferences, these laws offer a lens through which to address the challenges posed by the scalability of SR models. However, the presence of structural and collaborative issues in recommender systems prevents the direct application of the Scaling Law (SL) in these systems. In response, we introduce the Performance Law for SR models, which aims to theoretically investigate and model the relationship between model performance and data quality. Specifically, we first fit the HR and NDCG metrics to transformer-based SR models. Subsequently, we propose Approximate Entropy (ApEn) to assess data quality, presenting a more nuanced approach compared to traditional data quantity metrics. Our method enables accurate predictions across various dataset scales and model sizes, demonstrating a strong correlation in large SR models and offering insights into achieving optimal performance for any given model configuration.
format Preprint
id arxiv_https___arxiv_org_abs_2412_00430
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Optimizing Sequential Recommendation Models with Scaling Laws and Approximate Entropy
Shen, Tingjia
Wang, Hao
Wu, Chuhan
Chin, Jin Yao
Guo, Wei
Liu, Yong
Guo, Huifeng
Lian, Defu
Tang, Ruiming
Chen, Enhong
Artificial Intelligence
Information Retrieval
68P20
H.3.4; I.2.6
Scaling Laws have emerged as a powerful framework for understanding how model performance evolves as they increase in size, providing valuable insights for optimizing computational resources. In the realm of Sequential Recommendation (SR), which is pivotal for predicting users' sequential preferences, these laws offer a lens through which to address the challenges posed by the scalability of SR models. However, the presence of structural and collaborative issues in recommender systems prevents the direct application of the Scaling Law (SL) in these systems. In response, we introduce the Performance Law for SR models, which aims to theoretically investigate and model the relationship between model performance and data quality. Specifically, we first fit the HR and NDCG metrics to transformer-based SR models. Subsequently, we propose Approximate Entropy (ApEn) to assess data quality, presenting a more nuanced approach compared to traditional data quantity metrics. Our method enables accurate predictions across various dataset scales and model sizes, demonstrating a strong correlation in large SR models and offering insights into achieving optimal performance for any given model configuration.
title Optimizing Sequential Recommendation Models with Scaling Laws and Approximate Entropy
topic Artificial Intelligence
Information Retrieval
68P20
H.3.4; I.2.6
url https://arxiv.org/abs/2412.00430