Abstain Mask Retain Core: Time Series Prediction by Adaptive Masking Loss with Representation Consistency

Renzhao Liang, Sizhe Xu, Chenggang Xie, Jingru Chen, Feiyang Ren, Shu Yang, Takahiro Yabe · 2025

Read Abstain Mask Retain Core: Time Series Prediction by Adaptive Masking Loss with Representation Consistency by Renzhao Liang, Sizhe Xu, Chenggang Xie, Jingru Chen, Feiyang Ren, Shu Yang, Takahiro Yabe (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/arxiv-2510-19980.txt.

Description

Renzhao Liang and six co-authors challenge a basic assumption of time series forecasting: that more history is always better. Through careful experiments they find that trimming the past window can, oddly, improve accuracy. Models absorb redundant features, noise and irrelevant fluctuations, during training. Their remedy, AMRC, is built on information bottleneck thinking.

One component masks the training loss adaptively, spotlighting the most discriminative time segments to guide learning. A second enforces a consistency rule, stabilizing how inputs, labels, and predictions map to one another. Together they suppress redundant feature learning and significantly improve forecast accuracy across tests. The work offers forecasting practice both a caution and a concrete fix.

Catalog record

CREATORRenzhao Liang, Sizhe Xu, Chenggang Xie, Jingru Chen, Feiyang Ren, Shu Yang, Takahiro Yabe
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE83 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-19980

Browse the full library · About the Tower of Babel