Structural Autoencoders Improve Representations for Generation and Transfer

by   Felix Leeb, et al.

We study the problem of structuring a learned representation to significantly improve performance without supervision. Unlike most methods which focus on using side information like weak supervision or defining new regularization objectives, we focus on improving the learned representation by structuring the architecture of the model. We propose a self-attention based architecture to make the encoder explicitly associate parts of the representation with parts of the input observation. Meanwhile, our structural decoder architecture encourages a hierarchical structure in the latent space, akin to structural causal models, and learns a natural ordering of the latent mechanisms. We demonstrate how these models learn a representation which improves results in a variety of downstream tasks including generation, disentanglement, and transfer using several challenging and natural image datasets.


page 5

page 6

page 7

page 8


Reinforcement Learning with Neural Radiance Fields

It is a long-standing problem to find effective representations for trai...

VDSM: Unsupervised Video Disentanglement with State-Space Modeling and Deep Mixtures of Experts

Disentangled representations support a range of downstream tasks includi...

SALSA-TEXT : self attentive latent space based adversarial text generation

Inspired by the success of self attention mechanism and Transformer arch...

Pix2Shape – Towards Unsupervised Learning of 3D Scenes from Images using a View-based Representation

We infer and generate three-dimensional (3D) scene information from a si...

A New Modal Autoencoder for Functionally Independent Feature Extraction

Autoencoders have been widely used for dimensional reduction and feature...

Tile2Vec: Unsupervised representation learning for remote sensing data

Remote sensing lacks methods like the word vector representations and pr...

Adversarial Latent Autoencoder with Self-Attention for Structural Image Synthesis

Generative Engineering Design approaches driven by Deep Generative Model...

Please sign up or login with your details

Forgot password? Click here to reset