Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Kristian Kersting

Leveraging Probabilistic Circuits for Nonparametric Multi-Output Regression


Jun 16, 2021
Zhongjie Yu, Mingye Zhu, Martin Trapp, Arseny Skryagin, Kristian Kersting

* Accepted for the 37th Conference on Uncertainty in Artificial Intelligence (UAI 2021) 

  Access Paper or Ask Questions

RECOWNs: Probabilistic Circuits for Trustworthy Time Series Forecasting


Jun 08, 2021
Nils Thoma, Zhongjie Yu, Fabrizio Ventola, Kristian Kersting


  Access Paper or Ask Questions

Intriguing Parameters of Structural Causal Models


May 29, 2021
Matej Zečević, Devendra Singh Dhami, Kristian Kersting

* Main paper: 9 pages, References: 2 pages, Supplement: 2 pages. Main paper: 3 figures, Supplement: 1 figure 

  Access Paper or Ask Questions

User Label Leakage from Gradients in Federated Learning


May 22, 2021
Aidmar Wainakh, Fabrizio Ventola, Till Müßig, Jens Keim, Carlos Garcia Cordero, Ephraim Zimmer, Tim Grube, Kristian Kersting, Max Mühlhäuser


  Access Paper or Ask Questions

Decomposing 3D Scenes into Objects via Unsupervised Volume Segmentation


Apr 02, 2021
Karl Stelzner, Kristian Kersting, Adam R. Kosiorek

* 15 pages, 3 figures. For project page with videos, see http://stelzner.github.io/obsurf/ 

  Access Paper or Ask Questions

Language Models have a Moral Dimension


Mar 08, 2021
Patrick Schramowski, Cigdem Turan, Nico Andersen, Constantin Rothkopf, Kristian Kersting


  Access Paper or Ask Questions

Interventional Sum-Product Networks: Causal Inference with Tractable Probabilistic Models


Feb 23, 2021
Matej Zečević, Devendra Singh Dhami, Athresh Karanam, Sriraam Natarajan, Kristian Kersting

* Main paper: 8 pages, References: 2 pages, Appendix: 4 pages. Main paper: 4 figures, Appendix: 3 figures 

  Access Paper or Ask Questions

Recurrent Rational Networks


Feb 18, 2021
Quentin Delfosse, Patrick Schramowski, Alejandro Molina, Kristian Kersting

* Main paper: 8 pages, References: 2 pages, Appendix: 5 pages. Main paper: 6 figures, Appendix: 5 figures. Rational Activation Functions repository: https://github.com/ml-research/rational_activations Rational Supervised Learning: https://github.com/ml-research/rational_sl Rational Reinforcement Learning: https://github.com/ml-research/rational_rl 

  Access Paper or Ask Questions

Monte-Carlo Graph Search for AlphaZero


Dec 20, 2020
Johannes Czech, Patrick Korus, Kristian Kersting

* 11 pages, 7 figures, 3 tables 

  Access Paper or Ask Questions

Rule Extraction from Binary Neural Networks with Convolutional Rules for Model Validation


Dec 15, 2020
Sophie Burkhardt, Jannis Brugger, Nicolas Wagner, Zahra Ahmadi, Kristian Kersting, Stefan Kramer


  Access Paper or Ask Questions

Right for the Right Concept: Revising Neuro-Symbolic Concepts by Interacting with their Explanations


Dec 14, 2020
Wolfgang Stammer, Patrick Schramowski, Kristian Kersting


  Access Paper or Ask Questions

TUDataset: A collection of benchmark datasets for learning with graphs


Jul 16, 2020
Christopher Morris, Nils M. Kriege, Franka Bause, Kristian Kersting, Petra Mutzel, Marion Neumann

* ICML 2020 workshop "Graph Representation Learning and Beyond" 

  Access Paper or Ask Questions

Fitted Q-Learning for Relational Domains


Jun 10, 2020
Srijita Das, Sriraam Natarajan, Kaushik Roy, Ronald Parr, Kristian Kersting

* 10 pages, 12 figures 

  Access Paper or Ask Questions

Einsum Networks: Fast and Scalable Learning of Tractable Probabilistic Circuits


Apr 13, 2020
Robert Peharz, Steven Lang, Antonio Vergari, Karl Stelzner, Alejandro Molina, Martin Trapp, Guy Van den Broeck, Kristian Kersting, Zoubin Ghahramani


  Access Paper or Ask Questions

CryptoSPN: Privacy-preserving Sum-Product Network Inference


Feb 03, 2020
Amos Treiber, Alejandro Molina, Christian Weinert, Thomas Schneider, Kristian Kersting

* Accepted for publication at ECAI'20. Please cite the conference version of this paper 

  Access Paper or Ask Questions

Right for the Wrong Scientific Reasons: Revising Deep Networks by Interacting with their Explanations


Jan 31, 2020
Patrick Schramowski, Wolfgang Stammer, Stefano Teso, Anna Brugger, Xiaoting Shao, Hans-Georg Luigs, Anne-Katrin Mahlein, Kristian Kersting

* arXiv admin note: text overlap with arXiv:1805.08578 

  Access Paper or Ask Questions

BERT has a Moral Compass: Improvements of ethical and moral values of machines


Dec 11, 2019
Patrick Schramowski, Cigdem Turan, Sophie Jentzsch, Constantin Rothkopf, Kristian Kersting


  Access Paper or Ask Questions

Structured Object-Aware Physics Prediction for Video Modeling and Planning


Oct 06, 2019
Jannik Kossen, Karl Stelzner, Marcel Hussing, Claas Voelcker, Kristian Kersting


  Access Paper or Ask Questions

Neural Networks for Relational Data


Aug 28, 2019
Navdeep Kaur, Gautam Kunapuli, Saket Joshi, Kristian Kersting, Sriraam Natarajan

* 15 pages, 2 figures. To appear in the proceedings of 29th International Conference on Inductive Logic Programming 

  Access Paper or Ask Questions

Learning to play the Chess Variant Crazyhouse above World Champion Level with Deep Neural Networks and Human Data


Aug 22, 2019
Johannes Czech, Moritz Willig, Alena Beyer, Kristian Kersting, Johannes Fürnkranz

* 35 pages, 19 figures, 14 tables 

  Access Paper or Ask Questions

Random Sum-Product Forests with Residual Links


Aug 08, 2019
Fabrizio Ventola, Karl Stelzner, Alejandro Molina, Kristian Kersting


  Access Paper or Ask Questions

Padé Activation Units: End-to-end Learning of Flexible Activation Functions in Deep Networks


Jul 15, 2019
Alejandro Molina, Patrick Schramowski, Kristian Kersting

* 12 Pages, 6 Figures 

  Access Paper or Ask Questions

Declarative Learning-Based Programming as an Interface to AI Systems


Jun 18, 2019
Parisa Kordjamshidi, Dan Roth, Kristian Kersting


  Access Paper or Ask Questions

Neural-Symbolic Argumentation Mining: an Argument in Favour of Deep Learning and Reasoning


May 31, 2019
Andrea Galassi, Kristian Kersting, Marco Lippi, Xiaoting Shao, Paolo Torroni


  Access Paper or Ask Questions