Evaluating Disentanglement of Structured Latent Representations
- URL: http://arxiv.org/abs/2101.04041v1
- Date: Mon, 11 Jan 2021 17:24:01 GMT
- Title: Evaluating Disentanglement of Structured Latent Representations
- Authors: Rapha\"el Dang-Nhu and Angelika Steger
- Abstract summary: We design the first multi-layer disentanglement metric operating at all hierarchy levels of a structured latent representation.
Our metric unifies the evaluation of both object separation between latent slots and internal slot disentanglement into a common mathematical framework.
- Score: 3.756550107432323
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: We design the first multi-layer disentanglement metric operating at all
hierarchy levels of a structured latent representation, and derive its
theoretical properties. Applied to object-centric representations, our metric
unifies the evaluation of both object separation between latent slots and
internal slot disentanglement into a common mathematical framework. It also
addresses the problematic dependence on segmentation mask sharpness of previous
pixel-level segmentation metrics such as ARI. Perhaps surprisingly, our
experimental results show that good ARI values do not guarantee a disentangled
representation, and that the exclusive focus on this metric has led to
counterproductive choices in some previous evaluations. As an additional
technical contribution, we present a new algorithm for obtaining feature
importances that handles slot permutation invariance in the representation.
Related papers
- Measuring Orthogonality in Representations of Generative Models [81.13466637365553]
In unsupervised representation learning, models aim to distill essential features from high-dimensional data into lower-dimensional learned representations.
Disentanglement of independent generative processes has long been credited with producing high-quality representations.
We propose two novel metrics: Importance-Weighted Orthogonality (IWO) and Importance-Weighted Rank (IWR)
arXiv Detail & Related papers (2024-07-04T08:21:54Z) - Quantization of Large Language Models with an Overdetermined Basis [73.79368761182998]
We introduce an algorithm for data quantization based on the principles of Kashin representation.
Our findings demonstrate that Kashin Quantization achieves competitive or superior quality in model performance.
arXiv Detail & Related papers (2024-04-15T12:38:46Z) - Interpretability Benchmark for Evaluating Spatial Misalignment of
Prototypical Parts Explanations [13.111196926104485]
We name this undesired behavior a spatial explanation misalignment.
We propose a method for misalignment compensation and apply it to existing state-of-the-art models.
arXiv Detail & Related papers (2023-08-16T06:09:51Z) - 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) - Synergies between Disentanglement and Sparsity: Generalization and
Identifiability in Multi-Task Learning [79.83792914684985]
We prove a new identifiability result that provides conditions under which maximally sparse base-predictors yield disentangled representations.
Motivated by this theoretical result, we propose a practical approach to learn disentangled representations based on a sparsity-promoting bi-level optimization problem.
arXiv Detail & Related papers (2022-11-26T21:02:09Z) - Federated Representation Learning via Maximal Coding Rate Reduction [109.26332878050374]
We propose a methodology to learn low-dimensional representations from a dataset that is distributed among several clients.
Our proposed method, which we refer to as FLOW, utilizes MCR2 as the objective of choice, hence resulting in representations that are both between-class discriminative and within-class compressible.
arXiv Detail & Related papers (2022-10-01T15:43:51Z) - NashAE: Disentangling Representations through Adversarial Covariance
Minimization [8.22507807169023]
We present a self-supervised method to disentangle factors of variation in high-dimensional data that does not rely on prior knowledge of the underlying variation profile.
We show that NashAE has increased reliability and increased capacity to capture salient data characteristics in the learned latent representation.
arXiv Detail & Related papers (2022-09-21T22:02:26Z) - SGDR: Semantic-guided Disentangled Representation for Unsupervised
Cross-modality Medical Image Segmentation [5.090366802287405]
We propose a novel framework, called semantic-guided disentangled representation (SGDR), to exact semantically meaningful feature for segmentation task.
We validated our method on two public datasets and experiment results show that our approach outperforms the state of the art methods on two evaluation metrics by a significant margin.
arXiv Detail & Related papers (2022-03-26T08:31:00Z) - Dive into Ambiguity: Latent Distribution Mining and Pairwise Uncertainty
Estimation for Facial Expression Recognition [59.52434325897716]
We propose a solution, named DMUE, to address the problem of annotation ambiguity from two perspectives.
For the former, an auxiliary multi-branch learning framework is introduced to better mine and describe the latent distribution in the label space.
For the latter, the pairwise relationship of semantic feature between instances are fully exploited to estimate the ambiguity extent in the instance space.
arXiv Detail & Related papers (2021-04-01T03:21:57Z)
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.