Disentanglement Challenge: From Regularization to Reconstruction
This work addresses the problem of robust disentanglement in machine learning, but it is incremental as it builds on existing methods without introducing a new paradigm.
The authors tackled the challenge of learning disentangled representations by improving the FactorVAE method through enhanced reconstruction performance, increased network capacity, and more training steps, achieving first place in a competition using a real-world dataset.
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.