Jigsaw-VAE: Towards Balancing Features in Variational Autoencoders
arXiv:2005.05496
Abstract
The latent variables learned by VAEs have seen considerable interest as an unsupervised way of extracting features, which can then be used for downstream tasks. There is a growing interest in the question of whether features learned on one environment will generalize across different environments. We demonstrate here that VAE latent variables often focus on some factors of variation at the expense of others - in this case we refer to the features as ``imbalanced''. Feature imbalance leads to poor generalization when the latent variables are used in an environment where the presence of features changes. Similarly, latent variables trained with imbalanced features induce the VAE to generate less diverse (i.e. biased towards dominant features) samples. To address this, we propose a regularization scheme for VAEs, which we show substantially addresses the feature imbalance problem. We also introduce a simple metric to measure the balance of features in generated images.
References in corpus (17)
- mixup: Beyond Empirical Risk Minimization
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Equality of Opportunity in Supervised Learning
- Disentangling by Factorising
- Variational Lossy Autoencoder
- Neural Photo Editing with Introspective Adversarial Networks
- A Note on the Inception Score
- Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints
- Women also Snowboard: Overcoming Bias in Captioning Models
- Diagnosing and Enhancing VAE Models
- A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle
- Denoising Criterion for Variational Auto-Encoding Framework
- VAE with a VampPrior
- Disentangling Disentanglement in Variational Autoencoders
- Tackling Over-pruning in Variational Autoencoders
- Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases
- Jigsaw Puzzle Solving Using Local Feature Co-Occurrences in Deep Neural Networks