One Thing Leads to Another: The Essentials of Time Series Causal Discovery
A tutorial unifies time-series causal discovery methods, assumptions, evaluation, and uses in science, healthcare, and AI.
This tutorial unifies time-series causal discovery across signal processing, statistics, dynamical systems, and machine learning. It explains the principles, assumptions, and limitations of modern methods, plus how they are evaluated, benchmarked, and fine-tuned. It also discusses downstream uses in agriculture, Earth science, healthcare, and AI systems. A companion repository at CHAI-UK collects related code packages and resources.
- Surveys principles, assumptions, and limits of modern time-series causal discovery.
- Covers evaluation, benchmarking, and fine-tuning of discovery methods.
- Highlights uses in agriculture, Earth science, healthcare, and AI.
- A companion repository links code packages and learning resources.
- Organizations
- CHAI-UK
From the original article
1 min read at arxiv.orgTime series causal discovery --- the process of detecting causation between time series --- has its roots in signal processing research from over a half century ago. It has since received active and widespread interest, evolving into different forms across domains such as statistics, dynamical systems and machine learning. Entering this fragmented and constantly evolving research landscape can be daunting.
By Kurt Butler · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.