CausalVAE: Disentangled Representation Learning via Neural Structural Causal Models
Open accessThe DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
Abstract
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.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Causality: Models, Reasoning, and Inference
Pearl, J. (2009) · doi:10.1017/cbo9780511803161 · Cited by 11,653
Deep Learning Face Attributes in the Wild
Liu, Z., Luo, P., Wang, X., & Tang, X. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.425 · Cited by 7,747
(2017) · International Conference on Learning Representations · Cited by 3,248
Learning Bayesian Networks: The Combination of Knowledge and Statistical Data
Heckerman, D., Geiger, D., & Chickering, D. M. (1995) · Machine Learning · doi:10.1023/a:1022623210503 · Cited by 3,231
A Linear Non-Gaussian Acyclic Model for Causal Discovery
Shimizu, S., Hoyer, P. O., Hyvärinen, A., & Kerminen, A. (2006) · Cited by 2,261
DAGs with NO TEARS: Continuous Optimization for Structure Learning
Zheng, X., Aragam, B., Ravikumar, P., & Xing, E. P. (2018) · arXiv · doi:10.48550/arxiv.1803.01422 · Cited by 1,700 · cited 2×
“[34] proposed NOTEARs with a fully differentiable DAG constraint for causal structure learning, which drastically reduces a very complicated combinatorial optimization problem to a continuous optimization problem.”
BackgroundMethodDAG-GNN: DAG Structure Learning with Graph Neural NetworksPreprint
Yu, Y., Chen, J., Gao, T., & Yu, M. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.10098 · Cited by 728 · cited 3×
“The function attains 0 if and only if the adjacency matrix A corresponds to a DAG [32], i.e.”
BackgroundMethodChallenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
Locatello, F., Bauer, S., Lučić, M., Rätsch, G., Gelly, S., Schölkopf, B., & Bachem, O. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.12359 · Cited by 710 · cited 4×
“Unsupervised learning of the model might be infeasible due to the identifiability issue as discussed in [18].”
BackgroundInfluentialEquitability, mutual information, and the maximal information coefficient
Kinney, J. B., & Atwal, G. S. (2014) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1309933111 · Cited by 667 · cited 2×
“Table 1 records the mutual information (MIC/TIC) between the learned representation and the ground truth concept labels of all compared methods.”
MethodDeep Convolutional Inverse Graphics Network
Kulkarni, T. D., Whitney, W., Kohli, P., & Tenenbaum, J. B. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1503.03167 · Cited by 653 · cited 1×
“[15] and Locatello et al .”
Cited by
Robust and data-efficient generalization of self-supervised machine learning for diagnostic imaging
Azizi, S., Culp, L., Freyberg, J., Mustafa, B., Baur, S., Kornblith, S., Chen, T., Tomasev, N., Mitrović, J., Strachan, P., et al. (2023) · Nature Biomedical Engineering · doi:10.1038/s41551-023-01049-7 · Cited by 250
Towards Out-Of-Distribution Generalization: A Survey
Liu, J., Shen, Z., He, Y., Zhang, X., Xu, R., Yu, H., & Cui, P. (2021) · arXiv · doi:10.48550/arxiv.2108.13624 · Cited by 253
“…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],…”
BackgroundMethodInfluentialD’ya Like DAGs? A Survey on Structure Learning and Causal Discovery
Vowels, M. J., Camgoz, N. C., & Bowden, R. (2023) · ACM Computing Surveys · doi:10.1145/3527154 · Cited by 237
“This was undertaken in CausalVAE [257], who applied the NO TEARS constraint to a graph operating in low-dimensional representation space.”
BackgroundMethodInfluential
Metadata
- Authors
- Yang, M., Liu, F., Chen, Z., Shen, X., Hao, J., & Wang, J.
- Venue
- 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2021-06
- Pages
- 9588-9597
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W3038522363
- Cited by
- 249
- References
- 34
- FWCI
- 17.40
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.