Data Science and Intelligent Systems
New technologies and advances in data recording ranging from more and more satellites imaging earth, to handheld ultrasound devices, or automated wet lab experiments, lead to data creation at an unprecedented scale in nearly all disciplines. The high-dimensional and combinatorial nature as well as the sheer size of these datasets vastly exceeds human capabilities and the hope is that machine learning driven approaches can potentially aid in discovering or manipulating the data generating mechanism. Yet, even the most advanced deep learning based models are limited in their ability to understand the world around us, as shown in their poor exploration and generalization ability. Designing intelligent AI systems that generalize across environments and tasks in real-world settings is a fundamental, yet unsolved problem and our team aims to develop algorithms that learn causal relations from high-dimensional input, can explain their decisions by providing counterfactual answers and quickly adapt to new problems: key requirements for robust and transformational AI based enabling technologies in many application domains.
- N. Pfister, S. Bauer and J. Peters. "Identifying Causal Structure in Large-Scale Kinetic Systems“. Proceedings of the National Academy of Sciences (PNAS) (2019).
- B. Schölkopf, F. Locatello, S. Bauer, R. Nan Ke, N. Kalchbrenner, A. Goyal and Y. Bengio. "Towards Causal Representation Learning“. Proceedings of the IEEE - Advances in Machine Learning and Deep Neural Networks (2020).
- S. Bauer*, M. Wüthrich*, F. Widmaier*, N. Funk, J. De Jesus, J. Peters, J. Watson, C. Chen, K. Srinivasan, J. Zhang, J. Zhang, M. Walter, R. Madan, C. Schaff, T. Maeda, T. Yoneda, D. Yarats, A. Allshire, E. Gordon, T. Bhattacharjee, S. Srinivasa, A. Garg, A. Buchholz, S. Stark, T. Steinbrenner, J. Akpo, S. Joshi, V. Agrawal and B. Schölkopf. "A Robot Cluster in the Cloud for Reproducible Research in Dexterous Manipulation“. NeurIPS Competition Track (2021).
- P. Tigas, Y. Annadani, A. Jesson, B. Schölkopf, Y. Gal and S. Bauer. "Interventions, Where and How? Experimental Design for Causal Models at Scale“. Neural Information Processing Systems (NeurIPS) (2022).
- Z. Rao, P-Y. Tung, R. Xie, Y. Wei, H. Zhang, A. Ferrari, T. Klaver, F. Körmann, P. Sukumar, A. Kwiatkowski da Silva, Y. Chen, Z. Li, D. Ponge, J. Neugebauer, O. Gutfleisch, S. Bauer and D. Raabe. "Machine learning enabled fast high-entropy alloy discovery - a case study on novel INVAR alloys“. Science (2022).
Google scholar: Stefan Bauer