Targeted Reduction of Causal Models (2311.18639v2)
Abstract: Why does a phenomenon occur? Addressing this question is central to most scientific inquiries and often relies on simulations of scientific models. As models become more intricate, deciphering the causes behind phenomena in high-dimensional spaces of interconnected variables becomes increasingly challenging. Causal Representation Learning (CRL) offers a promising avenue to uncover interpretable causal patterns within these simulations through an interventional lens. However, developing general CRL frameworks suitable for practical applications remains an open challenge. We introduce Targeted Causal Reduction (TCR), a method for condensing complex intervenable models into a concise set of causal factors that explain a specific target phenomenon. We propose an information theoretic objective to learn TCR from interventional data of simulations, establish identifiability for continuous variables under shift interventions and present a practical algorithm for learning TCRs. Its ability to generate interpretable high-level explanations from complex models is demonstrated on toy and mechanical systems, illustrating its potential to assist scientists in the study of complex phenomena in a broad range of disciplines.
- Differentiable MPC for end-to-end planning and control. arXiv preprint arXiv:1810.13400, 2018.
- S. Beckers and J. Y. Halpern. Abstracting causal models. In Proceedings of the aaai conference on artificial intelligence, volume 33, pages 2678–2685, 2019.
- Approximate causal abstractions. In Uncertainty in artificial intelligence, pages 606–615. PMLR, 2020.
- M. Besserve and B. Schölkopf. Learning soft interventions in complex equilibrium systems. In Uncertainty in Artificial Intelligence, pages 170–180. PMLR, 2022.
- Counterfactuals uncover the modular structure of deep generative models. In ICLR2020, 2020.
- A discussion on euler method: A review. Electronic Journal of Mathematical Analysis and Applications, 1(2):2090–2792, 2013.
- Foundations of structural causal models with cycles and latent variables. The Annals of Statistics, 49(5):2885–2915, 2021.
- M. Breakspear. Dynamic models of large-scale brain activity. Nature neuroscience, 20(3):340–352, 2017.
- Causal structure-based root cause analysis of outliers. In International Conference on Machine Learning, pages 2357–2369. PMLR, 2022.
- Visual causal feature learning. arXiv preprint arXiv:1412.2309, 2014.
- Unsupervised discovery of el nino using causal feature learning on microlevel climate data. arXiv preprint arXiv:1605.09370, 2016.
- G. Deco and M. L. Kringelbach. Great expectations: using whole-brain computational connectomics for understanding neuropsychiatric disorders. Neuron, 84(5):892–905, 2014.
- Causal abstractions of neural networks. Advances in Neural Information Processing Systems, 34:9574–9586, 2021.
- Causal abstraction for faithful model interpretation. arXiv preprint arXiv:2301.04709, 2023a.
- Finding alignments between interpretable causal variables and distributed neural representations. arXiv preprint arXiv:2303.02536, 2023b.
- P. Geiger and C.-N. Straehle. Learning game-theoretic models of multiagent trajectories using implicit layers. arXiv preprint arXiv:2008.07303, 2020.
- H. Grassl. Status and improvements of coupled general circulation models. Science, 288(5473):1991–1997, 2000.
- Structured variable selection with sparsity-inducing norms. The Journal of Machine Learning Research, 12:2777–2824, 2011.
- Causal component analysis. arXiv preprint arXiv:2305.17225, 2023.
- Causal abstraction with soft interventions. In 2nd Conference on Causal Learning and Reasoning, 2023.
- A. Millet and P.-L. Morien. On implicit and explicit discretization schemes for parabolic spdes in any dimension. Stochastic Processes and their Applications, 115(7):1073–1106, 2005.
- J. Pearl. Causality. Cambridge university press, second edition, 2009.
- Elements of Causal Inference – Foundations and Learning Algorithms. MIT Press, 2017.
- Deep learning and process understanding for data-driven earth system science. Nature, 566(7743):195–204, 2019.
- E. F. Rischel and S. Weichwald. Compositional abstraction error and a category of causal models. In Uncertainty in Artificial Intelligence, pages 1013–1023. PMLR, 2021.
- Causal consistency of structural equation models. arXiv preprint arXiv:1707.00819, 2017.
- Uncovering the organization of neural circuits with generalized phase locking analysis. PLOS Computational Biology, 19(4):e1010983, 2023.
- T. Sauer. Computational solution of stochastic differential equations. Wiley Interdisciplinary Reviews: Computational Statistics, 5(5):362–371, 2013.
- Linear causal disentanglement via interventions. In 40th International Conference on Machine Learning, 2023.
- Nonparametric identifiability of causal representations from unknown interventions. arXiv preprint arXiv:2306.00542, 2023.
- Backtracking counterfactuals. In Conference on Causal Learning and Reasoning, pages 177–196. PMLR, 2023.
- Jointly learning consistent causal abstractions over multiple interventional distributions. arXiv preprint arXiv:2301.05893, 2023.
- Meaningful causal aggregation and paradoxical confounding. arXiv preprint arXiv:2304.11625, 2023.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.