Inducing Interpretable Representations with Variational Autoencoders
arXiv:1611.07492
Abstract
We develop a framework for incorporating structured graphical models in the \emph{encoders} of variational autoencoders (VAEs) that allows us to induce interpretable representations through approximate variational inference. This allows us to both perform reasoning (e.g. classification) under the structural constraints of a given graphical model, and use deep generative models to deal with messy, high-dimensional domains where it is often difficult to model all the variation. Learning in this framework is carried out end-to-end with a variational objective, applying to both unsupervised and semi-supervised schemes.
Presented at NIPS 2016 Workshop on Interpretable Machine Learning in Complex Systems
References in corpus (7)
- Adam: A Method for Stochastic Optimization
- Auto-Encoding Variational Bayes
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
- Deep Convolutional Inverse Graphics Network
- Attend, Infer, Repeat: Fast Scene Understanding with Generative Models
- Composing graphical models with neural networks for structured representations and fast inference