- Conference Paper
Rights / licenseIn Copyright - Non-Commercial Use Permitted
Disentanglement is at the forefront of unsupervised learning, as disentangled representations of data improve generalization, interpretability, and performance in downstream tasks. Current unsupervised approaches remain inapplicable for real-world datasets since they are highly variable in their performance and fail to reach levels of disentanglement of (semi-)supervised approaches. We introduce population-based training (PBT) for improving consistency in training variational autoencoders (VAEs) and demonstrate the validity of this approach in a supervised setting (PBT-VAE). We then use Unsupervised Disentanglement Ranking (UDR) as an unsupervised heuristic to score models in our PBT-VAE training and show how models trained this way tend to consistently disentangle only a subset of the generative factors. Building on top of this observation we introduce the recursive rPU-VAE approach. We train the model until convergence, remove the learned factors from the dataset and reiterate. In doing so, we can label subsets of the dataset with the learned factors and consecutively use these labels to train one model that fully disentangles the whole dataset. With this approach, we show striking improvement in state-of-the-art unsupervised disentanglement performance and robustness across multiple datasets and metrics. Show more
Book titleAdvances in Neural Information Processing Systems 33
Pages / Article No.
Organisational unit02533 - Institut für Neuroinformatik / Institute of Neuroinformatics
09474 - Yanik, Mehmet Fatih / Yanik, Mehmet Fatih
818179 - Engineering brain activity patterns for therapeutics of neuropsychiatric and neurological disorders (EC)
NotesPoster presentation held on December 8, 2020. Due to the Coronavirus (COVID-19) the conference was conducted virtually.
MoreShow all metadata