AI Paper Atlas

Research map / Time series / Causal and generative time series

Dynamic Time Warping Based Adversarial Framework for Time-Series Domain

Taha Belkhouja, Yan Yan, Janardhan Rao Doppa

TPAMI 2022 · Time series › Causal and generative time series · approach: State-space and latent dynamics forecasting · 61 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
use DTW-AR adversarial training instead of standard training1D-CNN · Heartbeatclean accuracy: 70% to 75%Improved
use DTW-AR adversarial training instead of l2-based attack training1D-CNN · UCR multivariateclean accuracyImproved
use random alignment paths instead of the single optimal DTW pathDTW-AR · five UCR datasetsfraction of dissimilar adversarial examples: DTW threshold 0.01: 100%Improved
set alpha_2 above zero (Euclidean-space penalty) (ablation)DTW-AR attack · UCR multivariatefooling rateImproved
replace Soft-DTW (CW-SDTW) with random-path DTW-ARDTW-AR · UCR multivariaterobustness / fooling rateImproved
use random-path distance instead of exact DTW per iterationDTW-AR · UCR multivariateruntime per iterationImproved
switch DTW variant from DTW_D to DTW_I (ablation)DTW-AR · UCR multivariateattack effectivenessNo clear effect
change norm in the DTW local distance among p=1,2,infinity (ablation)DTW-AR · UCR multivariateperformanceNo clear effect
train with resampling-warping augmentation instead of DTW-AR adversarial training1D-CNN · Epilepsy (DTW-AR adversarial examples)accuracy: 96 -> 26Hurt
train with resampling-warping augmentation instead of DTW-AR adversarial training1D-CNN · Heartbeat (DTW-AR adversarial examples)accuracy: 96 -> 20Hurt
use 1NN-DTW instead of 5NN or 10NN-DTW (ablation)kNN-DTW · five UCR datasetsaccuracy: Epilepsy 56 vs 40 and 32; Heartbeat 63 vs 67 and 68Mixed
use 1D-CNN instead of kNN-DTW on multivariate data1D-CNN · Epilepsyaccuracy: 56 -> 95Improved

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