Disentangling and Learning Robust Representations with Natural Clustering
arXiv:1901.09415 · doi:10.1109/ICMLA.2019.00125
Abstract
Learning representations that disentangle the underlying factors of variability in data is an intuitive way to achieve generalization in deep models. In this work, we address the scenario where generative factors present a multimodal distribution due to the existence of class distinction in the data. We propose N-VAE, a model which is capable of separating factors of variation which are exclusive to certain classes from factors that are shared among classes. This model implements an explicitly compositional latent variable structure by defining a class-conditioned latent space and a shared latent space. We show its usefulness for detecting and disentangling class-dependent generative factors as well as its capacity to generate artificial samples which contain characteristics unseen in the training data.
arXiv commentAccepted at ICMLA 2019
References in corpus (20)
- Auto-Encoding Variational Bayes
- Conditional Generative Adversarial Nets
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- Neural Discrete Representation Learning
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- Conditional Image Synthesis With Auxiliary Classifier GANs
- Semi-Supervised Learning with Deep Generative Models
- Improved Techniques for Training GANs
- InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
- Disentangling by Factorising
- Recent Advances in Autoencoder-Based Representation Learning
- Adversarial Autoencoders
- Understanding disentangling in -VAE
- DARLA: Improving Zero-Shot Transfer in Reinforcement Learning
- Structured Disentangled Representations
- Information Constraints on Auto-Encoding Variational Bayes
- DIVA: Domain Invariant Variational Autoencoders
- SCAN: Learning Hierarchical Compositional Visual Concepts
- Learning Disentangled Joint Continuous and Discrete Representations
- Auto-Encoding Total Correlation Explanation