Measuring Disentanglement: A Review of Metrics

• URL: http://arxiv.org/abs/2012.09276v2
• Date: Tue, 5 Jan 2021 22:37:16 GMT
• Title: Measuring Disentanglement: A Review of Metrics
• Authors: Julian Zaidi, Jonathan Boilard, Ghyslain Gagnon, Marc-Andr\'e Carbonneau
• Abstract summary: Learning to disentangle and represent factors of variation in data is an important problem in AI. We propose a new taxonomy in which all metrics fall into one of three families: intervention-based, predictor-based and information-based. We conduct extensive experiments, where we isolate representation properties to compare all metrics on many aspects.
• Score: 2.959278299317192
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Learning to disentangle and represent factors of variation in data is an important problem in AI. While many advances are made to learn these representations, it is still unclear how to quantify disentanglement. Several metrics exist, however little is known on their implicit assumptions, what they truly measure and their limits. As a result, it is difficult to interpret results when comparing different representations. In this work, we survey supervised disentanglement metrics and thoroughly analyze them. We propose a new taxonomy in which all metrics fall into one of three families: intervention-based, predictor-based and information-based. We conduct extensive experiments, where we isolate representation properties to compare all metrics on many aspects. From experiment results and analysis, we provide insights on relations between disentangled representation properties. Finally, we provide guidelines on how to measure disentanglement and report the results.

Related papers

• Understanding Probe Behaviors through Variational Bounds of Mutual Information [53.520525292756005]
  We provide guidelines for linear probing by constructing a novel mathematical framework leveraging information theory. First, we connect probing with the variational bounds of mutual information (MI) to relax the probe design, equating linear probing with fine-tuning. We show that the intermediate representations can have the biggest MI estimate because of the tradeoff between better separability and decreasing MI.
  arXiv  Detail & Related papers  (2023-12-15T18:38:18Z)
• An unsupervised learning approach to evaluate questionnaire data -- what one can learn from violations of measurement invariance [2.4762962548352467]
  This paper promotes an unsupervised learning-based approach to such research data. It works in three phases: data preparation, clustering of questionnaires, and measuring similarity based on the obtained clustering and the properties of each group. It provides a natural comparison between groups and a natural description of the response patterns of the groups.
  arXiv  Detail & Related papers  (2023-12-11T11:31:41Z)
• Enriching Disentanglement: From Logical Definitions to Quantitative Metrics [59.12308034729482]
  Disentangling the explanatory factors in complex data is a promising approach for data-efficient representation learning. We establish relationships between logical definitions and quantitative metrics to derive theoretically grounded disentanglement metrics. We empirically demonstrate the effectiveness of the proposed metrics by isolating different aspects of disentangled representations.
  arXiv  Detail & Related papers  (2023-05-19T08:22:23Z)
• On Quantitative Evaluations of Counterfactuals [88.42660013773647]
  This paper consolidates work on evaluating visual counterfactual examples through an analysis and experiments. We find that while most metrics behave as intended for sufficiently simple datasets, some fail to tell the difference between good and bad counterfactuals when the complexity increases. We propose two new metrics, the Label Variation Score and the Oracle score, which are both less vulnerable to such tiny changes.
  arXiv  Detail & Related papers  (2021-10-30T05:00:36Z)
• Disentanglement Analysis with Partial Information Decomposition [31.56299813238937]
  disentangled representations aim at reversing the process by mapping data to multiple random variables that individually capture distinct generative factors. Current disentanglement metrics are designed to measure the concentration, e.g., absolute deviation, variance, or entropy, of each variable conditioned by each generative factor. In this work, we use the Partial Information Decomposition framework to evaluate information sharing between more than two variables, and build a framework, including a new disentanglement metric.
  arXiv  Detail & Related papers  (2021-08-31T11:09:40Z)
• Ranking the information content of distance measures [61.754016309475745]
  We introduce a statistical test that can assess the relative information retained when using two different distance measures. This in turn allows finding the most informative distance measure out of a pool of candidates.
  arXiv  Detail & Related papers  (2021-04-30T15:57:57Z)
• Quantitative Evaluations on Saliency Methods: An Experimental Study [6.290238942982972]
  We briefly summarize the status quo of the metrics, including faithfulness, localization, false-positives, sensitivity check, and stability. We conclude that among all the methods we compare, no single explanation method dominates others in all metrics.
  arXiv  Detail & Related papers  (2020-12-31T14:13:30Z)
• Weakly-Supervised Disentanglement Without Compromises [53.55580957483103]
  Intelligent agents should be able to learn useful representations by observing changes in their environment. We model such observations as pairs of non-i.i.d. images sharing at least one of the underlying factors of variation. We show that only knowing how many factors have changed, but not which ones, is sufficient to learn disentangled representations.
  arXiv  Detail & Related papers  (2020-02-07T16:39:31Z)
• Disentangled Representation Learning with Wasserstein Total Correlation [90.44329632061076]
  We introduce Wasserstein total correlation in both variational autoencoder and Wasserstein autoencoder settings to learn disentangled latent representations. A critic is adversarially trained along with the main objective to estimate the Wasserstein total correlation term. We show that the proposed approach has comparable performances on disentanglement with smaller sacrifices in reconstruction abilities.
  arXiv  Detail & Related papers  (2019-12-30T05:31:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.