Disentanglement Challenge: From Regularization to Reconstruction

11/30/2019
by   Jie Qiao, et al.
0

The challenge of learning disentangled representation has recently attracted much attention and boils down to a competition using a new real world disentanglement dataset (Gondal et al., 2019). Various methods based on variational auto-encoder have been proposed to solve this problem, by enforcing the independence between the representation and modifying the regularization term in the variational lower bound. However recent work by Locatello et al. (2018) has demonstrated that the proposed methods are heavily influenced by randomness and the choice of the hyper-parameter. In this work, instead of designing a new regularization term, we adopt the FactorVAE but improve the reconstruction performance and increase the capacity of network and the training step. The strategy turns out to be very effective and achieve the 1st place in the challenge.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/30/2014

Auto-encoders: reconstruction versus compression

We discuss the similarities and differences between training an auto-enc...
research
05/20/2021

Comment on Stochastic Polyak Step-Size: Performance of ALI-G

This is a short note on the performance of the ALI-G algorithm (Berrada ...
research
06/20/2023

Variational Disentangled Graph Auto-Encoders for Link Prediction

With the explosion of graph-structured data, link prediction has emerged...
research
10/09/2018

Doubly Reparameterized Gradient Estimators for Monte Carlo Objectives

Deep latent variable models have become a popular model choice due to th...
research
04/30/2023

Learning Structured Output Representations from Attributes using Deep Conditional Generative Models

Structured output representation is a generative task explored in comput...
research
11/23/2020

The Connected Domination Number of Grids

Closed form expressions for the domination number of an n × m grid have ...
research
03/15/2020

Second order adjoint sensitivity analysis in variational data assimilation for tsunami models

We mathematically derive the sensitivity of data assimilation results fo...

Please sign up or login with your details

Forgot password? Click here to reset