TriMLP : A Foundational MLP-Like Architecture for Sequential Recommendation
ACM Transactions on Information Systems · 30.11.2024 · 42 (6) · 1-34
DOI: 10.1145/3670995 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Information Systems, online erschienen am 18.10.2024. Verfasst von Yiheng Jiang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 66 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of China for Young Scholars, Jilin Education Science Foundation.
Abstract (Originalsprache)
In this work, we present TriMLP as a foundational MLP-like architecture for the sequential recommendation, simultaneously achieving computational efficiency and promising performance. First, we empirically study the incompatibility between existing purely MLP-based models and sequential recommendation, that the inherent fully-connective structure endows historical user–item interactions (referred as tokens) with unrestricted communications and overlooks the essential chronological order in sequences. Then, we propose the MLP-based Triangular Mixer to establish ordered contact among tokens and excavate the primary sequential modeling capability under the standard auto-regressive training fashion. It contains (1) a global mixing layer that drops the lower-triangle neurons in MLP to block the anti-chronological connections from future tokens and (2) a local mixing layer that further disables specific upper-triangle neurons to split the sequence as multiple independent sessions. The mixer serially alternates these two layers to support fine-grained preferences modeling, where the global one focuses on the long-range dependency in the whole sequence, and the local one calls for the short-term patterns in sessions. Experimental results on 12 datasets of different scales from 4 benchmarks elucidate that TriMLP consistently attains favorable accuracy/efficiency tradeoff over all validated datasets, where the average performance boost against several state-of-the-art baselines achieves up to 14.88%, and the maximum reduction of inference time reaches 23.73%. The intriguing properties render TriMLP a strong contender to the well-established RNN-, CNN-, and Transformer-based sequential recommenders. Code is available at https://github.com/jiangyiheng1/TriMLP .
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of China for Young Scholars
- Jilin Education Science Foundation
- Jilin Science and Technology Research Project
- Jilin Science and Technology Development Project
- Science and Technology Development Fund (FDCT), Macau SAR
- Start-up Research Grant of University of Macau
Zitieren
APA
Jiang, Y., Xu, Y., Yang, Y., Yang, F., Wang, P., Li, C., Zhuang, F., & Xiong, H. (2024). TriMLP : A Foundational MLP-Like Architecture for Sequential Recommendation. ACM Transactions on Information Systems, 42(6), 1-34. https://doi.org/10.1145/3670995
Vancouver
Jiang Y, Xu Y, Yang Y, Yang F, Wang P, Li C, et al. TriMLP : A Foundational MLP-Like Architecture for Sequential Recommendation. ACM Transactions on Information Systems. 2024;42(6):1-34. doi:10.1145/3670995
BibTeX
@article{Jiang2024,
author = {Jiang, Yiheng and Xu, Yuanbo and Yang, Yongjian and Yang, Funing and Wang, Pengyang and Li, Chaozhuo and Zhuang, Fuzhen and Xiong, Hui},
title = {TriMLP : A Foundational MLP-Like Architecture for Sequential Recommendation},
journal = {ACM Transactions on Information Systems},
year = {2024},
volume = {42},
number = {6},
pages = {1--34},
doi = {10.1145/3670995}
}
EUVIMED