CausalVAE: Disentangled Representation Learning via Neural Structural Causal Models

By Mengyue Yang, Furui Liu, Zhitang Chen, Xinwei Shen, Jianye Hao, Jun Wang. 2020

407 citations recorded by the source corpus.

Learning disentanglement aims at finding a low dimensional representation which consists of multiple explanatory and generative factors of the observational data. The framework of variational autoencoder (VAE) is commonly used to disentangle independent factors from observations. However, in real scenarios, factors with semantics are not necessarily independent. Instead, there might be an underlying causal structure which renders these factors dependent. We thus propose a new VAE based framework named CausalVAE, which includes a Causal Layer to transform independent exogenous factors into causal endogenous ones that correspond to causally related concepts in data. We further analyze the model identifiabitily, showing that the proposed model learned from observations recovers the true one up to a certain degree. Experiments are conducted on various datasets, including synthetic and real word benchmark CelebA. Results show that the causal representations learned by CausalVAE are semantically interpretable, and their causal relationship as a Directed Acyclic Graph (DAG) is identified with good accuracy. Furthermore, we demonstrate that the proposed CausalVAE model is able to generate counterfactual data through "do-operation" to the causal factors.

Subfields: Artificial Intelligence, Generative Models, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Generative Models

In Representation & Self-Supervised Learning

Citations in the corpus

Disentanglement via Mechanism Sparsity Regularization: A New Principle for Nonlinear ICA (2021, background): “Yang et al. (2021) learns the causal graph by incorporating a “causal model layer” into iVAE, but does not rely on mechanism sparsity to disentangle, does not apply to time-series and is limited to linear CGMs.”
DiBS: Differentiable Bayesian Structure Learning (2021, background): “Follow-up work has leveraged this insight to model nonlinear relationships [24–27], time-series data [28], in the context of generative modeling [29, 30], for causal inference [31–33], and contributed to its theoretical understanding [34, 35].”
D’ya Like DAGs? A Survey on Structure Learning and Causal Discovery (2021, methodology): “This was undertaken in CausalVAE [257], who applied the NO TEARS constraint to a graph operating in low-dimensional representation space.”
Towards Out-Of-Distribution Generalization: A Survey (2021, methodology): “…Representation Learning ( Section 3.1 ) [27], [28], [29], [30], [31], [32], [33](Section 3.1) Causal Representation Learning ( Section 3.2 ) [34], [35], [36](Section 3.2) SupervisedLearningModels ( Section 4 ) ( Section 4.1 ) [7], [8], [9], [37], [38], [39], [40], [41], [42], [43], [44],…”