Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations

TLDR

Unsupervised disentangled representation learning assumes that data arise from a small number of explanatory factors that can be recovered without supervision. The paper critically examines recent progress in disentangled representation learning, challenges prevailing assumptions, and outlines future directions emphasizing inductive biases, supervision, and reproducibility. The authors conduct a reproducible large‑scale experiment training over 12,000 models across seven datasets to evaluate prominent disentanglement methods and metrics. The study demonstrates that unsupervised disentanglement is impossible without inductive biases, that methods cannot reliably produce identifiable disentangled models without supervision, and that higher disentanglement does not reduce sample complexity for downstream tasks.

Abstract

The key idea behind the unsupervised learning of disentangled representations is that real-world data is generated by a few explanatory factors of variation which can be recovered by unsupervised learning algorithms. In this paper, we provide a sober look at recent progress in the field and challenge some common assumptions. We first theoretically show that the unsupervised learning of disentangled representations is fundamentally impossible without inductive biases on both the models and the data. Then, we train more than 12000 models covering most prominent methods and evaluation metrics in a reproducible large-scale experimental study on seven different data sets. We observe that while the different methods successfully enforce properties ``encouraged'' by the corresponding losses, well-disentangled models seemingly cannot be identified without supervision. Furthermore, increased disentanglement does not seem to lead to a decreased sample complexity of learning for downstream tasks. Our results suggest that future work on disentanglement learning should be explicit about the role of inductive biases and (implicit) supervision, investigate concrete benefits of enforcing disentanglement of the learned representations, and consider a reproducible experimental setup covering several data sets.