News

The Causal Inference group is part of German Aerospace Center’s Institute of Data Science in Jena (website there). Since 2021 another branch of the group is located at my chair of Climate Informatics at Technische Universität (TU) Berlin as part of an ERC Starting Grant.

Current open positions in Jena (flexible home office rules also allow working from another city)

  • Postdoc / PhD position within DLR project “Causal Anomalies” on causal inference methods for causal attribution of anomalies in various application domains.
  • Postdoc position within DLR project “Causal Inference” on general causal inference theory and methods, as well as relations to explainable AI. The ad is coming soon, you can send an email (single PDF with CV, grades, motivation) to jakob.runge at dlr.de.
  • Postdoc position within ERC CausalEarth on application (!) of causal methods on climate observations and model data. This position is targeted at climate scientists with experience in CMIP models and basic knowledge of ML/statistics/causal inference.

In all positions you will develop theory and methods at the forefront of causal inference and AI and help climate and other domain scientists apply them to better understand complex dynamical systems such as the Earth. The well-funded positions offer ample opportunities for research travel and more extended stays at our international partners and can also be involved in the ELLIS network among leading European AI researchers. We look for applicants with a strong background in math/stats/physics/ML and especially encourage women to apply. A background in climate science is NOT needed.

Note on Home Office: At DLR remote work may be possible, in agreement with the employer, to the extent of up to a maximum of 80 percent of the contractually agreed work performance in a calendar month. This means that we are open to you working remotely from another (German) city and coming to Jena on a regular basis.

Open positions in Berlin

erc_logoThis is part of the ERC Starting Grant CausalEarth at Technical University (TU) Berlin where I am guest professor in the computer science department since February 2021. The new branch enables lively exchange and collaborations with the world-class TU research chairs on AI and machine learning. CausalEarth (more information here) develops causal inference and discovery methods inspired by challenges of the spatio-temporal complex system Earth.

New NeurIPS (2020) paper

Checkout our new NeurIPS paper on high-recall causal discovery for autocorrelated time series with latent confounders. Joint work with Andreas Gerhardus. More comparisons soon on the causality benchmark platform http://causeme.net

New UAI (2020) paper

Checkout my new UAI paper on time series causal discovery for lagged AND contemporaneous dependencies (PCMCI+). Often causal links have a shorter time lag than the resolution of the time series, which leads to contemporaneous links that cannot be directed using methods like Granger causality and also my previous method PCMCI (Science Advances paper below). PCMCI+ addresses this issue and can identify contemporaneous causal directions even between just two time series, if any of them has enough autocorrelation. Works very well also for many variables. More comparisons soon on the causality benchmark platform http://causeme.net

New Science Advances paper on causal discovery

Science_Advances_picIdentifying causal relationships and quantifying their strength from observational time series data are key problems in disciplines dealing with complex dynamical systems such as the Earth system. Data-driven causal inference in such systems is challenging since datasets are often high-dimensional and nonlinear with limited sample sizes. Here we introduce a novel method that flexibly combines linear or nonlinear conditional independence tests with a causal discovery algorithm to estimate causal networks from large-scale time series datasets. Our method outperforms state-of-the-art techniques in detection power, which opens up entirely new possibilities to discover and quantify causal networks from time series across a range of research fields. The paper is accompanied by the software package Tigramite. Also have a look at the Perspective paper and causality benchmark platform below!

Nature Communications perspective paper on causal discovery from time series

news_imageTogether with co-authors, we recently published a Nature Communications perspective paper on causal discovery from time series. The perspective provides an overview of causal inference methods, identifies promising applications, and discusses methodological challenges (exemplified in Earth system sciences). I hope you find it useful for your work!