AI Paper Atlas

Research map / Time series / Foundation models for time series

CALDA: Improving Multi-Source Time Series Domain Adaptation With Contrastive Adversarial Learning

Garrett Wilson, Janardhan Rao Doppa, Diane J. Cook

TPAMI 2023 · Time series › Foundation models for time series · approach: Time series representation learning · 69 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
add domain adversary (ablation)CALDA-Any,R · Real-world averageaccuracy: 73.2 -> 80.2Improved
add domain adversary (ablation)CALDA-XS,H · Real-world averageaccuracy: 73.1 -> 79.9Improved
add pseudo-labeled target data to contrastive loss (ablation)CALDA-Any,R · Real-world averageaccuracy: 80.2 -> 76.2Hurt
add pseudo-labeled target data to contrastive loss (ablation)CALDA-XS,H · Real-world averageaccuracy: 79.9 -> 75.4Hurt
use cross-source/any-source pairs instead of within-source pairs (ablation)CALDA · Synthetic datasets with shiftaccuracy: 71.5 -> 75.0Improved
hard sampling instead of random sampling (ablation)CALDA-Any · Real-world averageaccuracy: 80.2 -> 79.7Hurt
hard sampling instead of random sampling (ablation)CALDA-Any · Synthetic datasets with shiftaccuracy: 71.5 -> 75.0Improved
hard sampling instead of random sampling (ablation)CALDA-XS · WISDM ARaccuracy: 80.0 -> 81.4Improved
add contrastive learning plus adversary over CoDATSCALDA-Any,R · Real-world averageaccuracy: 78.6 -> 80.2Improved
add weak supervision regularizerCALDA-Any,R · Real-world averageaccuracy: 80.2 -> 83.7Improved
add weak supervision regularizerCALDA-XS,H · WISDM ATaccuracy: 71.0 -> 83.3Improved
add label-proportion noise budget 0.4 to weak supervisionCoDATS-WS · WISDM ATaccuracy: 72.1 -> 56.3Hurt

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