Unsupervised and interpretable scene discovery with Discrete-Attend-Infer-Repeat
arXiv:1903.06581
Abstract
In this work we present Discrete Attend Infer Repeat (Discrete-AIR), a Recurrent Auto-Encoder with structured latent distributions containing discrete categorical distributions, continuous attribute distributions, and factorised spatial attention. While inspired by the original AIR model andretaining AIR model's capability in identifying objects in an image, Discrete-AIR provides direct interpretability of the latent codes. We show that for Multi-MNIST and a multiple-objects version of dSprites dataset, the Discrete-AIR model needs just one categorical latent variable, one attribute variable (for Multi-MNIST only), together with spatial attention variables, for efficient inference. We perform analysis to show that the learnt categorical distributions effectively capture the categories of objects in the scene for Multi-MNIST and for Multi-Sprites.
References in corpus (9)
- Adam: A Method for Stochastic Optimization
- Categorical Reparameterization with Gumbel-Softmax
- InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
- DRAW: A Recurrent Neural Network For Image Generation
- Disentangling by Factorising
- Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding
- Attend, Infer, Repeat: Fast Scene Understanding with Generative Models
- Isolating Sources of Disentanglement in Variational Autoencoders
- Synthesizing Programs for Images using Reinforced Adversarial Learning