AI Paper Atlas

Research map / Time series / Foundation models for time series

TimeGuard: Channel-wise Pool Training for Backdoor Defense in Time Series Forecasting

Quang Duc Nguyen, Siyuan Liang, Yiming Li, Fushuo Huo, Dacheng Tao

ICML 2026 · Time series › Foundation models for time series · approach: Time series representation learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
adopt channel-wise reliable pool training instead of sample-level filtering (ablation)TimeGuard · PEMS03, Random attackFDER: 0.478 -> 0.868Improved
add Distance-Regularized Loss Selection instead of loss-only selection (ablation)TimeGuard · PEMS03, Random attackFDER: 0.607 -> 0.868Improved
add Neighborhood Diversity Filtering to Stage I pool initialization (ablation)TimeGuard · PEMS03, Random attackFDER: 0.853 -> 0.868Improved
add Reverse-Consistency Filtering to Stage I pool initialization (ablation)TimeGuard · PEMS03, BackTime attackFDER: 0.796 -> 0.808Improved
add NDF and RCF together in Stage I (ablation)TimeGuard · PEMS03, Random attackMAEP: 91.780 -> 104.677Improved
use TimeGuard instead of the strongest baseline PDBPEMS03 averaged over attacks · PEMS03MAEP: 1.96x relative improvement in MAEPImproved
apply sample-level filtering defenses (Spectral, TED, TED++) to TSFSpectral, TED, TED++ · PEMS03FDER: FDER about 0.54No clear effect
rely on training-loss-only defenses (ABL, ESTI)ABL, ESTI · PEMS03FDER: average FDER 0.497No clear effect
use inference-time detection (STRIP, TeCo, IBD-PSC) adapted to TSFTeCo (best) · PEMS03, Weather, ETTm1AUROC: 0.551 AUROC and 0.559 F1 for TeCoNo clear effect
set neighbourhood size K from small to large (ablation)TimeGuard · PEMS03, BackTimeFDER: FDER 0.805-0.809No clear effect
increase scaling factor pi beyond 1.5 (ablation)TimeGuard · PEMS03, BackTimeFDERHurt
set pool size beta too small or too large (best 0.5-0.7) (ablation)TimeGuard · PEMS03, BackTimeFDER: MAEP 39.28 at beta 0.7 vs 33.59 at beta 0.8 (alpha 0.25)Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy