preprints and publications

2022

1. Evaluating Robustness to Dataset Shift via Parametric Robustness Sets

   Thams, Nikolaj*, Oberst, Michael*, and Sontag, David

   *Equal contribution, order determined by coin flip

   In Advances in Neural Information Processing Systems (NeurIPS) 2022

   arXiv Bib Code

   @inproceedings{thams2022evaluating,
     title = {Evaluating Robustness to Dataset Shift via Parametric Robustness Sets},
     author = {Thams, Nikolaj* and Oberst, Michael* and Sontag, David},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
     volume = {36},
     arxiv = {2205.15947},
     year = {2022},
     note = {*Equal contribution, order determined by coin flip},
     code = {https://github.com/clinicalml/parametric-robustness-evaluation}
   }

2. Identifying Causal Effects using Instrumental Time Series: Nuisance IV and Correcting for the Past

   Thams, Nikolaj, Søndergaard, Rikke, Weichwald, Sebastian, and Peters, Jonas

   arXiv preprint arXiv:2203.06056 2022

   arXiv Bib Code

   @article{thams2022identifying,
     title = {Identifying Causal Effects using Instrumental Time Series: Nuisance IV and Correcting for the Past},
     author = {Thams, Nikolaj and Søndergaard, Rikke and Weichwald, Sebastian and Peters, Jonas},
     journal = {arXiv preprint arXiv:2203.06056},
     arxiv = {2203.06056},
     year = {2022},
     code = {https://github.com/nikolajthams/its-time}
   }

3. Invariant Ancestry Search

   Mogensen, Phillip,  Thams, Nikolaj, and Peters, Jonas

   In Proceedings of the 39th International Conference on Machine Learning 17–23 jul 2022

   arXiv Bib HTML Code

   @inproceedings{mogensen2022invariant,
     title = {Invariant Ancestry Search},
     author = {Mogensen, Phillip and Thams, Nikolaj and Peters, Jonas},
     booktitle = {Proceedings of the 39th International Conference on Machine Learning},
     volume = {162},
     pages = {15832--15857},
     series = {Proceedings of Machine Learning Research},
     month = {17--23 Jul},
     publisher = {PMLR},
     arxiv = {2202.00913},
     year = {2022},
     html = {https://proceedings.mlr.press/v162/mogensen22a.html},
     code = {https://github.com/PhillipMogensen/InvariantAncestrySearch}
   }

2021

1. Local Independence Testing for Point Processes

   Thams, Nikolaj, and Hansen, Niels Richard

   arXiv preprint arXiv:2110.12709 2021

   arXiv Bib Code

   @article{thams2021local,
     title = {Local Independence Testing for Point Processes},
     author = {Thams, Nikolaj and Hansen, Niels Richard},
     journal = {arXiv preprint arXiv:2110.12709},
     arxiv = {2110.12709},
     year = {2021},
     code = {https://github.com/nikolajthams/LIPP}
   }

2. Invariant Policy Learning: A Causal Perspective

   Saengkyongam, Sorawit,  Thams, Nikolaj, Peters, Jonas, and Pfister, Niklas

   arXiv preprint arXiv:2106.00808 2021

   arXiv Bib Code

   @article{saengkyongam2021invariant,
     title = {Invariant Policy Learning: A Causal Perspective},
     author = {Saengkyongam, Sorawit and Thams, Nikolaj and Peters, Jonas and Pfister, Niklas},
     journal = {arXiv preprint arXiv:2106.00808},
     arxiv = {2106.00808},
     year = {2021},
     code = {https://github.com/sorawitj/invariant-policy-learning}
   }

3. Statistical Testing under Distributional Shifts

   Thams, Nikolaj, Saengkyongam, Sorawit, Pfister, Niklas, and Peters, Jonas

   Journal of the Royal Statistical Society, Series B 2021

   arXiv Bib Code

   @article{thams2021statistical,
     title = {Statistical Testing under Distributional Shifts},
     author = {Thams, Nikolaj and Saengkyongam, Sorawit and Pfister, Niklas and Peters, Jonas},
     journal = {Journal of the Royal Statistical Society, Series B},
     volume = {(To appear)},
     arxiv = {2105.10821},
     year = {2021},
     code = {https://github.com/nikolajthams/testing-under-shifts}
   }

4. Regularizing towards Causal Invariance: Linear Models with Proxies

   Oberst, Michael,  Thams, Nikolaj, Peters, Jonas, and Sontag, David

   In Proceedings of the 38th International Conference on Machine Learning 18–24 jul 2021

   arXiv Bib HTML Code

   @inproceedings{oberst2021regularizing,
     title = {Regularizing towards Causal Invariance: Linear Models with Proxies},
     author = {Oberst, Michael and Thams, Nikolaj and Peters, Jonas and Sontag, David},
     booktitle = {Proceedings of the 38th International Conference on Machine Learning},
     pages = {8260--8270},
     year = {2021},
     editor = {Meila, Marina and Zhang, Tong},
     volume = {139},
     series = {Proceedings of Machine Learning Research},
     month = {18--24 Jul},
     publisher = {PMLR},
     html = {http://proceedings.mlr.press/v139/oberst21a.html},
     arxiv = {2103.02477},
     code = {https://github.com/clinicalml/proxy-anchor-regression}
   }

2020

1. Causal structure learning from time series: Large regression coefficients may predict causal links better in practice than small p-values

   Weichwald, Sebastian, Jakobsen, Martin E, Mogensen, Phillip B, Petersen, Lasse,  Thams, Nikolaj, and Varando, Gherardo

   In NeurIPS 2019 Competition and Demonstration Track 2020

   arXiv Bib HTML Code

   @inproceedings{weichwald2020causal,
     title = {Causal structure learning from time series: Large regression coefficients may predict causal links better in practice than small p-values},
     author = {Weichwald, Sebastian and Jakobsen, Martin E and Mogensen, Phillip B and Petersen, Lasse and Thams, Nikolaj and Varando, Gherardo},
     booktitle = {NeurIPS 2019 Competition and Demonstration Track},
     pages = {27--36},
     year = {2020},
     html = {http://proceedings.mlr.press/v123/weichwald20a/weichwald20a.pdf},
     arxiv = {2002.09573},
     organization = {PMLR},
     code = {https://github.com/sweichwald/tidybench}
   }

2019

1. Thams, N. (2019). Master thesis: Causal structure learning in multivariate point processes.

   Bib PDF

   @thesis{thams2019causal,
     title = {Master thesis: Causal structure learning in multivariate point processes},
     author = {Thams, Nikolaj},
     pdf = {master_thesis.pdf},
     year = {2019}
   }