AI Paper Atlas

Research map / Time series / Foundation models for time series

AutoDA-Timeseries: Automated Data Augmentation for Time Series

Zijun Dou, Zhenhe Yao, Zhe Xie, Xidao Wen, Tong Xiao, Dan Pei

ICLR 2026 · Time series › Foundation models for time series · approach: Time series classification

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
add time series features to condition augmentation policy (ablation)AutoDA-Timeseries (RNN) · Long-term forecasting, ExchangeMSE: 0.4301 -> 0.3409Improved
add time series features (ablation)AutoDA-Timeseries (RNN) · Long-term forecasting, ETTh1MSE: 0.5612 -> 0.5429Improved
add time series features (ablation)AutoDA-Timeseries (RNN) · Long-term forecasting, WeatherMSE: 0.2689 -> 0.2574Improved
add joint optimization of probabilities and intensities (ablation)AutoDA-Timeseries (RNN) · Long-term forecasting, ExchangeMSE: 0.4395 -> 0.3409Improved
add dynamic temperature (ablation)AutoDA-Timeseries (RNN) · Long-term forecasting, ExchangeMSE: 0.4235 -> 0.3409Improved
add composite loss (ablation)AutoDA-Timeseries (RNN) · Long-term forecasting, ExchangeMSE: 0.4322 -> 0.3409Improved
include autocorrelation Catch22 features (ablation)AutoDA-Timeseries · UEA classification EPaccuracy: 0.9275 -> 0.9783Improved
include differencing Catch22 features (ablation)AutoDA-Timeseries · UEA classification RSaccuracy: 0.8026 -> 0.8947Improved
initialize with uniform distribution plus raw bonus instead of pure uniform (ablation)AutoDA-Timeseries · UEA classification (ROCKET)average accuracy: 0.9187 -> 0.9353Improved
apply UniformAugment instead of no augmentation in cross-dataset transferNoAug · ETTh1 to ETTh2 transfer (RNN)MSE: 0.5290 -> 0.7188Hurt
apply AutoDA-Timeseries instead of no augmentation in cross-dataset transferNoAug · ETTh1 to ETTh2 transfer (RNN)MSE: 0.5290 -> 0.5076Improved
apply TS2Vec representation learning instead of no augmentationNoAug · UEA classification (TCN)accuracy: 0.685 -> 0.584Hurt

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