Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modeling
arXiv:2010.13187
Abstract
Current autoencoder-based disentangled representation learning methods achieve disentanglement by penalizing the (aggregate) posterior to encourage statistical independence of the latent factors. This approach introduces a trade-off between disentangled representation learning and reconstruction quality since the model does not have enough capacity to learn correlated latent variables that capture detail information present in most image data. To overcome this trade-off, we present a novel multi-stage modeling approach where the disentangled factors are first learned using a penalty-based disentangled representation learning method; then, the low-quality reconstruction is improved with another deep generative model that is trained to model the missing correlated latent variables, adding detail information while maintaining conditioning on the previously learned disentangled factors. Taken together, our multi-stage modelling approach results in a single, coherent probabilistic model that is theoretically justified by the principal of D-separation and can be realized with a variety of model classes including likelihood-based models such as variational autoencoders, implicit models such as generative adversarial networks, and tractable models like normalizing flows or mixtures of Gaussians. We demonstrate that our multi-stage model has higher reconstruction quality than current state-of-the-art methods with equivalent disentanglement performance across multiple standard benchmarks. In addition, we apply the multi-stage model to generate synthetic tabular datasets, showcasing an enhanced performance over benchmark models across a variety of metrics. The interpretability analysis further indicates that the multi-stage model can effectively uncover distinct and meaningful features of variations from which the original distribution can be recovered.
References in corpus (18)
- Auto-Encoding Variational Bayes
- Conditional Generative Adversarial Nets
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Instance Normalization: The Missing Ingredient for Fast Stylization
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- Generative Adversarial Text to Image Synthesis
- NICE: Non-linear Independent Components Estimation
- Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
- Density estimation using Real NVP
- Disentangling by Factorising
- Invertible Conditional GANs for image editing
- VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning
- Disentangling factors of variation in deep representations using adversarial training
- Learning in Implicit Generative Models
- Are Disentangled Representations Helpful for Abstract Visual Reasoning?
- Variational Inference of Disentangled Latent Concepts from Unlabeled Observations
- BreGMN: scaled-Bregman Generative Modeling Networks