Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence 2022
DOI: 10.24963/ijcai.2022/297
|View full text |Cite
|
Sign up to set email alerts
|

MLP4Rec: A Pure MLP Architecture for Sequential Recommendations

Abstract: We address the efficiency problem of personalized ranking from implicit feedback by hashing users and items with binary codes, so that top-N recommendation can be fast executed in a Hamming space by bit operations. However, current hashing methods for top-N recommendation fail to align their learning objectives (such as pointwise or pairwise loss) with the benchmark metrics for ranking quality (e.g. Average Precision, AP), resulting in sub-optimal accuracy. To this end, we propose a Discrete Listwise Personali… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
5
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
5
4
1

Relationship

2
8

Authors

Journals

citations
Cited by 28 publications
(5 citation statements)
references
References 0 publications
0
5
0
Order By: Relevance
“…• LightGCN linearly propagates information on the useritem interaction graph by simplified graph convolutions to learn high-order connectivity (He et al 2020). • MLP4Rec uses MLP-Mixer (Tolstikhin et al 2021) as the backbone and proposes a tri-directional fusion scheme for sequential recommendation (Li et al 2022).…”
Section: Compared Baselinesmentioning
confidence: 99%
“…• LightGCN linearly propagates information on the useritem interaction graph by simplified graph convolutions to learn high-order connectivity (He et al 2020). • MLP4Rec uses MLP-Mixer (Tolstikhin et al 2021) as the backbone and proposes a tri-directional fusion scheme for sequential recommendation (Li et al 2022).…”
Section: Compared Baselinesmentioning
confidence: 99%
“…The purpose of sequential recommender systems is to address the order of interactions [22,38]. Early methods are mainly based on Markov Chain and Markov Processes.…”
Section: Sequential Recommender Systemsmentioning
confidence: 99%
“…MOI-Mixer [13] first investigates the possibility of using MLP architecture as a substitute for Transformer-based methods. Subsequently, MLP4Rec [16] proposed a tri-directional information fusion scheme, to coherently capture higher-order interactions across different attribute levels under a feature-rich scenario.…”
Section: Mlp-mixermentioning
confidence: 99%