DBScholar

Back to papers

A Multi-Scale Decomposition MLP-Mixer for Time Series Analysis

Summary: MSD-Mixer: an MLP‑Mixer that explicitly decomposes uni-/multivariate time series using multi-scale temporal patching and MLPs to model intra/inter-patch dynamics and channel correlations. New loss constrains residual mean and autocorrelation for decomposition completeness; achieves consistent, efficient SOTA across five time-series tasks. (summarized by gpt-5-mini on Feb 09 2026)

Paper ID
13599
Venue
VLDB
Year
2024
Pagerank
5.3483178e-05
Overall Rank
8,915 | 38.84%
DOI
10.14778/3654621.3654637

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@article{zhong_vldb24,
        title = {{A Multi-Scale Decomposition MLP-Mixer for Time Series Analysis}},
        author = {Zhong, Shuhan and Song, Sizhe and Zhuo, Weipeng and Li, Guanyao and Liu, Yang and Chan, S.-H. Gary},
        journal = {PVLDB},
        series = {{VLDB} '24},
        volume = {17},
        number = {7},
        pages = {1723--1736},
        doi = {10.14778/3654621.3654637},
        url = {https://doi.org/10.14778/3654621.3654637},
        year = {2024}
}

Incoming Citations (Sorted by Pagerank)

Showing 1 of 1 citing papers.

Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 11 of 11 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Previous Page 1 / 1 Next

Semantically Similar Papers