Causal discovery in climate research: Overview and recent progress
Bibliographic record
Abstract
Discovering causal dependencies from observational time series datasets is a major problem in better understanding the complex dynamical system Earth. Recent methodological advances have addressed major challenges such as high-dimensionality and nonlinearity (PCMCI, Runge et al. Sci. Adv. 2019), as well as instantaneous causal links (PCMCI+, Runge UAI, 2020) and hidden variables (LPCMCI, Gerhardus and Runge, 2020), but many more remain. In this presentation I will give an overview of challenges and methods and present a recent approach, Ensemble-PCMCI, to analyze ensembles of climate time series. An example for this are initialized ensemble forecasts. Since the individual samples can then be created from several time series instead of different time steps from a single time series, such cases allow to relax the assumption of stationarity and hence to analyze whether and how the underlying causal relationships change over time. We compare Ensemble-PCMCI to other methods and discuss preliminary applications. Runge et al., Detecting and quantifying causal associations in large nonlinear time series datasets, Science Advances eeaau4996 (2019). Runge, J. Discovering contemporaneous and lagged causal relations in autocorrelated nonlinear time series datasets. Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence, UAI 2020, Toronto, Canada, 2019, AUAI Press, 2020 Gerhardus, A. & Runge, J. High-recall causal discovery for autocorrelated time series with latent confounders. Advances in Neural Information Processing Systems, 2020, 33
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.037 | 0.053 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.003 |
| Bibliometrics | 0.010 | 0.016 |
| Science and technology studies | 0.001 | 0.006 |
| Scholarly communication | 0.007 | 0.011 |
| Open science | 0.004 | 0.005 |
| Research integrity | 0.005 | 0.007 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".