Disentangling by Subspace Diffusion
This work addresses the challenge of unsupervised representation learning for researchers in machine learning and differential geometry, providing a foundational insight that reduces disentangling to metric learning.
The paper tackles the problem of unsupervised disentangling of data manifolds by introducing the Geometric Manifold Component Estimator (GEOMANCER), a nonparametric algorithm that factors manifolds based on symmetry and holonomy, such as rotation in 3D, without requiring labeled data.
We present a novel nonparametric algorithm for symmetry-based disentangling of data manifolds, the Geometric Manifold Component Estimator (GEOMANCER). GEOMANCER provides a partial answer to the question posed by Higgins et al. (2018): is it possible to learn how to factorize a Lie group solely from observations of the orbit of an object it acts on? We show that fully unsupervised factorization of a data manifold is possible if the true metric of the manifold is known and each factor manifold has nontrivial holonomy -- for example, rotation in 3D. Our algorithm works by estimating the subspaces that are invariant under random walk diffusion, giving an approximation to the de Rham decomposition from differential geometry. We demonstrate the efficacy of GEOMANCER on several complex synthetic manifolds. Our work reduces the question of whether unsupervised disentangling is possible to the question of whether unsupervised metric learning is possible, providing a unifying insight into the geometric nature of representation learning.