Timezone: »
While probabilistic models are an important tool for studying causality, doing so suffers from the intractability of inference. As a step towards tractable causal models, we consider the problem of learning interventional distributions using sum-product networks (SPNs) that are over-parameterized by gate functions, e.g., neural networks. Providing an arbitrarily intervened causal graph as input, effectively subsuming Pearl's do-operator, the gate function predicts the parameters of the SPN. The resulting interventional SPNs are motivated and illustrated by a structural causal model themed around personal health. Our empirical evaluation against competing methods from both generative and causal modelling demonstrates that interventional SPNs indeed are both expressive and causally adequate.
Author Information
Matej Zečević (TU Darmstadt)
##### You can check out my personal website, with cool articles (also non-scientific!): [www.matej-zecevic.de](https://www.matej-zecevic.de)
Devendra Dhami (CS Department, TU Darmstadt, TU Darmstadt)
Athresh Karanam (University of Texas, Dallas)
Sriraam Natarajan (Indiana University)
Kristian Kersting (TU Darmstadt)
More from the Same Authors
-
2021 : Deep RePReL--Combining Planning and Deep RL for acting in relational domains »
Harsha Kokel · Arjun Manoharan · Sriraam Natarajan · Balaraman Ravindran · Prasad Tadepalli -
2022 : Mixture of Gaussian Processes with Probabilistic Circuits for Multi-Output Regression »
Mingye Zhu · Zhongjie Yu · Martin Trapp · Arseny Skryagin · Kristian Kersting -
2023 : LEDITS++: Limitless Image Editing using Text-to-Image Models »
Manuel Brack · Linoy Tsban · Katharina Kornmeier · Apolinário Passos · Felix Friedrich · Patrick Schramowski · Kristian Kersting -
2023 : LEDITS++: Limitless Image Editing using Text-to-Image Models »
Manuel Brack · Linoy Tsban · Katharina Kornmeier · Apolinário Passos · Felix Friedrich · Patrick Schramowski · Kristian Kersting -
2023 : Leveraging Diffusion-Based Image Variations for Robust Training on Poisoned Data »
Lukas Struppek · Martin Bernhard Hentschel · Clifton Poth · Dominik Hintersdorf · Kristian Kersting -
2023 : Defending Our Privacy With Backdoors »
Dominik Hintersdorf · Lukas Struppek · Daniel Neider · Kristian Kersting -
2023 Poster: Do Not Marginalize Mechanisms, Rather Consolidate! »
Moritz Willig · Matej Zečević · Devendra Dhami · Kristian Kersting -
2023 Poster: Interpretable and Explainable Logical Policies via Neurally Guided Symbolic Abstraction »
Quentin Delfosse · Hikaru Shindo · Devendra Dhami · Kristian Kersting -
2023 Poster: ATMAN: Understanding Transformer Predictions Through Memory Efficient Attention Manipulation »
Björn Deiseroth · Mayukh Deb · Samuel Weinbach · Manuel Brack · Patrick Schramowski · Kristian Kersting -
2023 Poster: SEGA: Instructing Text-to-Image Models using Semantic Guidance »
Manuel Brack · Felix Friedrich · Dominik Hintersdorf · Lukas Struppek · Patrick Schramowski · Kristian Kersting -
2023 Poster: MultiFusion: Fusing Pre-Trained Models for Multi-Lingual, Multi-Modal Image Generation »
Marco Bellagente · Manuel Brack · Hannah Teufel · Felix Friedrich · Björn Deiseroth · Constantin Eichenberg · Andrew Dai · Robert Baldock · Souradeep Nanda · Koen Oostermeijer · Andres Felipe Cruz-Salinas · Patrick Schramowski · Kristian Kersting · Samuel Weinbach -
2023 Poster: Characteristic Circuits »
Zhongjie Yu · Martin Trapp · Kristian Kersting -
2023 Oral: Characteristic Circuits »
Zhongjie Yu · Martin Trapp · Kristian Kersting -
2022 : Closing Remarks »
Matej Zečević -
2022 : Panel Discussion: "Heading for a Unifying View on nCSI" »
Tobias Gerstenberg · Sriraam Natarajan · - Mausam · Guy Van den Broeck · Devendra Dhami -
2022 Workshop: Workshop on neuro Causal and Symbolic AI (nCSI) »
Matej Zečević · Devendra Dhami · Christina Winkler · Thomas Kipf · Robert Peharz · Petar Veličković -
2022 : Welcome & Opening Remarks »
Matej Zečević -
2022 : Panel »
Guy Van den Broeck · Cassio de Campos · Denis Maua · Kristian Kersting · Rianne van den Berg -
2022 Poster: ORIENT: Submodular Mutual Information Measures for Data Subset Selection under Distribution Shift »
Athresh Karanam · Krishnateja Killamsetty · Harsha Kokel · Rishabh Iyer -
2017 Tutorial: Statistical Relational Artificial Intelligence: Logic, Probability and Computation »
Luc De Raedt · David Poole · Kristian Kersting · Sriraam Natarajan