Hi, I am Sebastian Tschiatschek and currently assistant professor (tenure-track) of machine learning at the University of Vienna. Before that I was a senior researcher in the Machine Learning and Perception Group (now All Data AI) at Microsoft Research Cambridge. Before joining Microsoft I was a postdoc in the Learning and Adaptive Systems Group at ETH Zurich led by Prof. Andreas Krause. Before joining ETH, I received a PhD from Graz University of Technology under supervision of Prof. Franz Pernkopf at the Signal and Speech Processing Laboratory.
My research interests include probabilistic models (with a focus on set-valued inputs and outputs) and exploration and imitiaton in reinforcement learning.
I am looking for two PhD students: one doing research on topics in reinforcement learning and another one doing research on topics related to interpretable and explainable machine learning in the context of democracy and decision making. If you are interested, please contact me directly. Information on the first position is available here, information on the latter position is available here.
Paper Accepted at IJCAI'21
One paper was accepted at IJCAI:
Lukas Miklautz, Lena Greta-Marie Bauer, Dominik Mautz, Sebastian Tschiatschek, Christian Böhm, Claudia Plant, “Details (Don’t) Matter: Isolating Cluster Information in Deep Embedded Spaces”
Congrats Lukas and Lena!
Tutorial at IJCAI
Adish Singla and I will give a tutorial on Recent Advances in Reinforcement Learning for Human-AI Collaboration at this year’s IJCAI. Details will follow.
Workshop proposal accepted at ICML
Our workshop on Human-AI Collaboration in Sequential Decision-Making was accepted at this year’s ICML. Please consider the call for papers.
Papers Accepted at AAAI and EAAI
One paper was accepted at AAAI and one at EAAI:
- Zichao Wang, Sebastian Tschiatschek, Simon Woodhead, Jose Miguel Hernandez Lobato, Simon Peyton Jones, Richard Baraniuk, Cheng Zhang, “Educational Question Mining At Scale: Prediction, Analysis and Personalization”
- Haiyan Yin, Jianda Chen, Sinno Pan, Sebastian Tschiatschek, “Sequential Generative Exploration Model for Partially Observable Reinforcement Learning”
Paper Accepted at NeurIPS
One paper accepted at this year’s NeurIPS:
- Chao Ma, Sebastian Tschiatschek, Richard E Turner, José Miguel Hernández-Lobato, Cheng Zhang, “VAEM: a Deep Generative Model for Heterogeneous Mixed Type Data”
Lecture/Tutorial on Machine Learning
- I gave a leture/tutorial on Machine Learning at the Pattern Recognition in Neuroimaging Summer School 2020.
New Preprints on Arxiv
Two new preprints are available on arxiv:
- Dongge Han, Shruti Tople, Alex Rogers, Michael Wooldridge, Olga Ohrimenko, Sebastian Tschiatschek, “Replication-Robust Payoff-Allocation with Applications in Machine Learning Marketplaces”
- Chao Ma, Sebastian Tschiatschek, José Miguel Hernández-Lobato, Richard Turner, Cheng Zhang, VAEM: a Deep Generative Model for Heterogeneous Mixed Type Data
Visual agents for augmenting people’s own capabilities
Here’s an interesting article on a project I was involved in while being at Microsoft Research Cambridge: Link
I have joined the University of Vienna as assistant professor (tenure-track) of machine learning.
Our paper was accepted at AAMAS 2020:
- Ahana Ghosh, Sebastian Tschiatschek, Hamed Mahdavi, Adish Singla, “Towards Deployment of Robust Cooperative AI Agents: An Algorithmic Framework for Learning Adaptive Policies”
The following paper was accepted at ICLR 2020:
- Jacob Beck, Kamil Ciosek, Sam Devlin, Sebastian Tschiatschek, Cheng Zhang, Katja Hofmann, “AMRL: Aggregated Memory For Reinforcement Learning” [Video presentation]
We have four papers accepted at NeurIPS 2019!
- Sebastian Tschiatschek, Ahana Ghosh, Luis Haug, Rati Devidze, Adish Singla, “Learner-aware Teaching: Inverse Reinforcement Learning with Preferences and Constraints”
- David Janz, Jiri Hron, Przemysław Mazur, Katja Hofmann, Jose Miguel Hernandez-Lobato, Sebastian Tschiatschek, “Successor Uncertainties: Exploration and Uncertainty in Temporal Difference Learning”
- Maximilian Igl, Kamil Ciosek, Yingzhen Li, Sebastian Tschiatschek, Cheng Zhang, Sam Devlin, Katja Hofmann, “Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck”
- Wenbo Gong, Sebastian Tschiatschek, Sebastian Nowozin, Richard E Turner, Jose Miguel Hernandez-Lobato, Cheng Zhang, “Icebreaker: Efficient Information Acquisition with Active Learning”
Please check for updates of the papers from time to time as these are not yet the final versions.
We have one paper at the main conference and one paper in the Real-world Sequential Decision Making: Reinforcement Learning and Beyond workshop:
- Chao Ma, Sebastian Tschiatschek, Konstantina Palla, José Miguel Hernández-Lobato, Sebastian Nowozin, Cheng Zhang, “EDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAE”
Talk: Wed Jun 12th 11:40 AM — 12:00 PM @ Hall A
Poster: Wed Jun 12th 6:30 PM @ Pacific Ballroom
- Chao Ma, Cheng Zhang, Sebastian Tschiatschek, Jose Miguel Hernandez-Lobato, Richard E. Turner, Sebastian Nowozin, “Bayesian EDDI: Sequential Variable Selection with Bayesian Partial VAE”
Poster: Fri Jun 14th, Real-world Sequential Decision Making: Reinforcement Learning and Beyond Workshop
We have one paper at the main conference and one paper in the deep reinforcement learning workshop:
- Luis Haug, Sebastian Tschiatschek, and Adish Singla, “Teaching Inverse Reinforcement Learners via Features and Demonstrations” Poster: Wed Dec 5th 05:00 – 07:00 PM @ Room 517 AB #167
- David Janz, Jiri Hron, José Miguel Hernández-Lobato, Katja Hofmann, and Sebastian Tschiatschek, “Successor Uncertainties: exploration and uncertainty in temporal difference learning” Poster: Fri Dec 7th, Deep RL Workshop