Corpus ID: 222140982

Disentangled Generative Causal Representation Learning

@article{Shen2020DisentangledGC,
  title={Disentangled Generative Causal Representation Learning},
  author={Xinwei Shen and Furui Liu and Hanze Dong and Qing Lian and Zhitang Chen and Tong Zhang},
  journal={ArXiv},
  year={2020},
  volume={abs/2010.02637}
}
This paper proposes a Disentangled gEnerative cAusal Representation (DEAR) learning method. Unlike existing disentanglement methods that enforce independence of the latent variables, we consider the general case where the underlying factors of interests can be causally correlated. We show that previous methods with independent priors fail to disentangle causally correlated factors. Motivated by this finding, we propose a new disentangled learning method called DEAR that enables causal… Expand
2 Citations
D'ya like DAGs? A Survey on Structure Learning and Causal Discovery
  • Highly Influenced
  • PDF

References

SHOWING 1-10 OF 47 REFERENCES
CausalVAE: Disentangled Representation Learning via Neural Structural Causal Models
  • 7
  • PDF
CausalVAE: Structured Causal Disentanglement in Variational Autoencoder
  • 5
Variational Inference of Disentangled Latent Concepts from Unlabeled Observations
  • 204
  • PDF
CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training
  • 69
  • Highly Influential
  • PDF
InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANs
  • 15
  • PDF
Disentangling Factors of Variation Using Few Labels
  • 43
  • Highly Influential
  • PDF
A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms
  • 101
  • PDF
Weakly Supervised Disentanglement with Guarantees
  • 30
  • PDF
Weakly-Supervised Disentanglement Without Compromises
  • 31
  • Highly Influential
  • PDF
...
1
2
3
4
5
...