Learning latent causal graphs via mixture oracles
arXiv:2106.15563
Abstract
We study the problem of reconstructing a causal graphical model from data in the presence of latent variables. The main problem of interest is recovering the causal structure over the latent variables while allowing for general, potentially nonlinear dependence between the variables. In many practical problems, the dependence between raw observations (e.g. pixels in an image) is much less relevant than the dependence between certain high-level, latent features (e.g. concepts or objects), and this is the setting of interest. We provide conditions under which both the latent representations and the underlying latent causal model are identifiable by a reduction to a mixture oracle. These results highlight an intriguing connection between the well-studied problem of learning the order of a mixture model and the problem of learning the bipartite structure between observables and unobservables. The proof is constructive, and leads to several algorithms for explicitly reconstructing the full graphical model. We discuss efficient algorithms and provide experiments illustrating the algorithms in practice.
arXiv commentTo appear at NeurIPS 2021. 41 pages
References in corpus (21)
- Identifiability of parameters in latent structure models with many observed variables
- Latent variable graphical model selection via convex optimization
- On the Identifiability of the Post-Nonlinear Causal Model
- Identifiability of Gaussian structural equation models with equal error variances
- Pixels to Graphs by Associative Embedding
- Causal Inference in the Presence of Latent Variables and Selection Bias
- Data spectroscopy: Eigenspaces of convolution operators and clustering
- CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training
- Graphs for margins of Bayesian networks
- Nested Markov Properties for Acyclic Directed Mixed Graphs
- Causal Discovery in Physical Systems from Videos
- Markovian acyclic directed mixed graphs for discrete data
- Margins of discrete Bayesian networks
- Learning loopy graphical models with latent variables: Efficient methods and guarantees
- On Multi-Cause Causal Inference with Unobserved Confounding: Counterexamples, Impossibility, and Alternatives
- Finite state space non parametric Hidden Markov Models are in general identifiable
- Naïve regression requires weaker assumptions than factor models to adjust for multiple cause confounding
- A generalization of Kruskal's theorem on tensor decomposition
- Consistent Estimation of Identifiable Nonparametric Mixture Models from Grouped Observations
- GANs with Conditional Independence Graphs: On Subadditivity of Probability Divergences
- Efficient Tensor Decomposition