Julius von Kügelgen
Exploring the intersection of causal inference and machine learning
MPI for Intelligent Systems / W. Scheible
I am a PhD student with Bernhard Schölkopf at the Max Planck Institute for Intelligent Systems in Tübingen. As part of the Cambridge-Tübingen programme, I am also co-supervised by Adrian Weller at the University of Cambridge, where I spent the first year of my PhD.
My research interests lie at the intersection of causal inference and machine learning, see Research Interests for more details.
Previously, I studied Mathematics (BSc+MSci) at Imperial College London and Artificial Intelligence (MSc) at UPC Barcelona in Spain and at TU Delft in the Netherlands.
I am originally from the beautiful Hamburg in northern Germany.
06/2022 - I will talk about Active Bayesian Causal Inference at the International Society for Bayesian Analysis (ISBA) World Meeting in Montreal, as part of an invited session on "Bayesian experimental design for causal inference"; if you're also at ISBA, feel free to reach out ;)
05/2022 - I gave a lecture on "Introduction to Causal Inference" at the Gdańsk Machine Learning Summer School; here are my slides.
05/2022 - "Causal Inference Through the Structural Causal Marginal Problem" accepted at ICML 2022, co-led with Luigi Gresele & Jonas Kübler.
04/2022 - Together with Amir Karimi, we've written a book chapter "Toward Causal Algorithmic Recourse" as part of the Springer LNAI 13200 "xxAI - Beyond Explainable AI".
04/2022 - Together with Bernhard Schölkopf, we've written a review "From Statistical to Causal Learning", to appear in the Proceeding of the International Congress of Mathematicians 2022.
03/2022 - We are organising the 1st Workshop on Causal Representation Learning at UAI'22 in Eindhoven, The Netherlands on 5 August (in-person/hybrid). Check out the website for details about the workshop and a list of selected references.
03/2022 - I have started my academic visit(s) in the US, where I will be for the next four months:
03/2022 - "Complex interlinkages, key objectives and nexuses amongst the Sustainable Development Goals and climate change: a network analysis" accepted at The Lancet Planetary Health, led by the excellent Felix Laumann.
01/2022 - With a great team of co-authors, two papers accepted at ICLR 2022:
12/2021 - I gave an invited talk at the WHY-21 NeurIPS Workshop "Causal Inference & Machine Learning: Why now?": links to slides and recording.
12/2021 - "On the Fairness of Causal Algorithmic Recourse" accepted at AAAI 2022 (oral).
09/2021 - With amazing co-authors, three papers accepted at NeurIPS 2021:
08/2021 - I will help organise the 1st Conference on Causal Learning and Reasoning (CLeaR) as part of the Logistics and Conference Planning Team.
08/2021 - "Causal Direction of Data Collection Matters: Implications of Causal and Anticausal Learning for NLP" accepted at EMNLP 2021 (oral).
Selected publications & preprints
On the Fairness of Causal Algorithmic Recourse.
AAAI 2022 (oral)
(Previously: ICML 2021 Workshop Algorithmic Recourse; NeurIPS 2020 Workshop Algorithmic Fairness through the Lens of Causality and Interpretability (AFCI) )
Algorithmic recourse under imperfect causal knowledge: a probabilistic approach
NeurIPS 2020 (spotlight; top 4% of submissions)
(Previously: ICML 2020 Workshops
XXAI: Extending Explainable AI Beyond Deep Models and Classifiers (oral; 4 out of 20 papers)
WHI: Workshop on Human Interpretability in Machine Learning (oral; 4 out of 50 papers))
Semi-supervised learning, causality and the conditional cluster assumption.
JvK, Alexander Mey, Marco Loog, Bernhard Schölkopf.
(Previously: NeurIPS 2019 Workshop “Do the right thing”: machine learning and causal inference for improved decision making)
Other publications & preprints
On the DCI framework for evaluating disentangled representations: Extensions and connections to identifiability
Cian Eastwood*, Andrei Liviu Nicolicioiu*, JvK*, Armin Kekic, Frederik Träuble, Andrea Dittadi, Bernhard Schölkopf. (*equal contribution)
Complex interlinkages, key objectives and nexuses amongst the Sustainable Development Goals and climate change: a network analysis
Felix Laumann, JvK, Thiago Hector Kanashiro Uehara, Mauricio Barahona
The Lancet Planetary Health
Visual Representation Learning Does Not Generalize Strongly Within the Same Domain.
Lukas Schott, JvK, Frederik Träuble, Peter Gehler, Chris Russell, Matthias Bethge, Bernhard Schölkopf, Francesco Locatello, Wieland Brendel.
(Previously at: ICLR 2021 Workshop Generalization beyond the training distribution in brains and machines )
Non-linear interlinkages and key objectives amongst the Paris Agreement and the Sustainable Development Goals.
Felix Laumann, JvK, Mauricio Barahona.
ICLR 2020 Workshop Tackling Climate Change with Machine Learning (CCAI)
Optimal experimental design via Bayesian optimisation: active causal structure learning for Gaussian process networks.
JvK, Paul K Rubenstein, Bernhard Schölkopf, Adrian Weller.
NeurIPS 2019 Workshop “Do the right thing”: machine learning and causal inference for improved decision making