[2203.16437] Weakly supervised causal representation learning
Learning high-level causal representations together with a causal model from unstructured low-level data such as pixels is impossible from observational data alone. We prove under mild assumptions that this representation is however identifiable in a weakly supervised setting. This involves a dataset with paired samples before and after random, unknown interventions, but no further labels. We then introduce implicit latent causal models, variational autoencoders that represent causal variables and causal structure without having to optimize an explicit discrete graph structure. On simple image data, including a novel dataset of simulated robotic manipulation, we demonstrate that such models can reliably identify the causal structure and disentangle causal variables.
View PDF HTML (experimental) Abstract:Learning high-level causal representations together with a causal model from unstructured low-level data such as pixels is impossible from observational data alone. We prove under mild assumptions that this representation is however identifiable in a weakly supervised setting. This involves a dataset with paired samples before and after random, unknown interventions, but no further labels. We then introduce implicit latent causal models, variational autoencoders that represent causal variables and causal structure without having to optimize an explicit…
saved by
related reading
- Faithful, Interpretable Model Explanations via Causal Abstraction | SAIL Blogai.stanford.edu
- Causal Scrubbing: a method for rigorously testing interpretability hypotheses [Redwood Research] — LessWronglesswrong.com
- Tutorial on Variational Autoencodersarxiv.org
- Causal Autoregressive Flowsarxiv.org
- How Causal Abstraction Underpins Computational Explanationarxiv.org
- [1906.02691] An Introduction to Variational Autoencodersarxiv.org
- 2006.10742arxiv.org
- [2210.13382] Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Taskarxiv.org
- Imitating Latent Policies from Observationarxiv.org
- Relating Graph Neural Networks to Structural Causal Modelsarxiv.org
- Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Taskarxiv.org
- Tutorial - What is a variational autoencoder? – Jaan Lı 李jaan.io