DCI-ES: An Extended Disentanglement Framework with Connections to Identifiability
This work provides incremental improvements to evaluation metrics for disentangled representation learning, primarily benefiting researchers in representation learning and independent component analysis.
The authors extended the DCI framework for evaluating disentangled representations by adding two new measures (explicitness and size) and connecting it to identifiability concepts, demonstrating relevance on MPI3D and Cars3D datasets.
In representation learning, a common approach is to seek representations which disentangle the underlying factors of variation. Eastwood & Williams (2018) proposed three metrics for quantifying the quality of such disentangled representations: disentanglement (D), completeness (C) and informativeness (I). In this work, we first connect this DCI framework to two common notions of linear and nonlinear identifiability, thereby establishing a formal link between disentanglement and the closely-related field of independent component analysis. We then propose an extended DCI-ES framework with two new measures of representation quality - explicitness (E) and size (S) - and point out how D and C can be computed for black-box predictors. Our main idea is that the functional capacity required to use a representation is an important but thus-far neglected aspect of representation quality, which we quantify using explicitness or ease-of-use (E). We illustrate the relevance of our extensions on the MPI3D and Cars3D datasets.