Meta Variational Memory Transformer for Anomaly Detection of Multivariate Time Series
Kun Qin, Yuxin Li, Wenchao Chen, Xinyue Hu, Bo Chen, Hongwei Liu · 2025
Read Meta Variational Memory Transformer for Anomaly Detection of Multivariate Time Series by Kun Qin, Yuxin Li, Wenchao Chen, Xinyue Hu, Bo Chen, Hongwei Liu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12736823.txt.
Description
Qin and five co-authors built a memory transformer in twenty twenty-five to spot odd runs in many-channel time data. Stock nets fix one map per stream, which costs much and bends poor-ly to new streams. Their net codes stream kinds into shared memory slots first.
Those slots then seed a maker model that picks fit priors per stream. A two-way guess step reads the hidden states of each stream. On six data sets the net led one-for-all odd-spot tasks. Shared memory gives the flex that fixed maps lack, they show.
Catalog record
| CREATOR | Kun Qin, Yuxin Li, Wenchao Chen, Xinyue Hu, Bo Chen, Hongwei Liu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |