Invariant Representations from Adversarially Censored Autoencoders
arXiv:1805.08097
Abstract
We combine conditional variational autoencoders (VAE) with adversarial censoring in order to learn invariant representations that are disentangled from nuisance/sensitive variations. In this method, an adversarial network attempts to recover the nuisance variable from the representation, which the VAE is trained to prevent. Conditioning the decoder on the nuisance variable enables clean separation of the representation, since they are recombined for model learning and data reconstruction. We show this natural approach is theoretically well-founded with information-theoretic arguments. Experiments demonstrate that this method achieves invariance while preserving model learning performance, and results in visually improved performance for style transfer and generative sampling tasks.
arXiv comment9 pages, 5 figures
References in corpus (9)
- Auto-Encoding Variational Bayes
- Adversarial Feature Learning
- Adversarially Learned Inference
- Adversarial Discriminative Domain Adaptation
- Adversarial Autoencoders
- Disentangling factors of variation in deep representations using adversarial training
- Censoring Representations with an Adversary
- Learning to Pivot with Adversarial Networks
- Controllable Invariance through Adversarial Feature Learning
Cited by in corpus (5)
- Fairness in Machine Learning
- Representation via Representations: Domain Generalization via Adversarially Learned Invariant Representations
- Learning Fair and Transferable Representations
- Generative Deep Learning Model for a Multi-level Nano-Optic Broadband Power Splitter
- Transfer Learning in Brain-Computer Interfaces with Adversarial Variational Autoencoders