Corpus ID: 211082941

Learning Flat Latent Manifolds with VAEs

@inproceedings{Chen2020LearningFL,
  title={Learning Flat Latent Manifolds with VAEs},
  author={Nutan Chen and Alexej Klushyn and F. Ferroni and J. Bayer and P. V. D. Smagt},
  booktitle={ICML},
  year={2020}
}
Measuring the similarity between data points often requires domain knowledge, which can in parts be compensated by relying on unsupervised methods such as latent-variable models, where similarity/distance is estimated in a more compact latent space. Prevalent is the use of the Euclidean metric, which has the drawback of ignoring information about similarity of data stored in the decoder, as captured by the framework of Riemannian geometry. We propose an extension to the framework of variational… Expand
5 Citations
Encoded Prior Sliced Wasserstein AutoEncoder for learning latent manifold representations
  • Highly Influenced
  • PDF
On Implicit Regularization in β-VAEs
  • 7
  • Highly Influenced
  • PDF
Variational Autoencoders for Learning Nonlinear Dynamics of Physical Systems
  • PDF

References

SHOWING 1-10 OF 60 REFERENCES
Simple online and realtime tracking with a deep association metric
  • 660
  • Highly Influential
  • PDF
mixup: Beyond Empirical Risk Minimization
  • 1,631
  • Highly Influential
  • PDF
beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework
  • 1,779
  • Highly Influential
Ladder Variational Autoencoders
  • 482
  • Highly Influential
  • PDF
Stochastic Backpropagation and Approximate Inference in Deep Generative Models
  • 3,127
  • Highly Influential
  • PDF
Learning Hierarchical Priors in VAEs
  • 26
  • PDF
Active Learning based on Data Uncertainty and Model Sensitivity
  • 7
  • PDF
Metrics for Deep Generative Models
  • 41
  • PDF
Change Detection in Graph Streams by Learning Graph Embeddings on Constant-Curvature Manifolds
  • 12
  • PDF
Fast Approximate Geodesics for Deep Generative Models
  • 14
  • PDF
...
1
2
3
4
5
...