Attribute Based Interpretable Evaluation Metrics for Generative Models

• URL: http://arxiv.org/abs/2310.17261v3
• Date: Wed, 17 Jul 2024 14:13:08 GMT
• Title: Attribute Based Interpretable Evaluation Metrics for Generative Models
• Authors: Dongkyun Kim, Mingi Kwon, Youngjung Uh,
• Abstract summary: We propose a new evaluation protocol that measures the divergence of a set of generated images from the training set regarding the distribution of attribute strengths. Our metrics lay a foundation for explainable evaluations of generative models.
• Score: 14.407813583528968
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: When the training dataset comprises a 1:1 proportion of dogs to cats, a generative model that produces 1:1 dogs and cats better resembles the training species distribution than another model with 3:1 dogs and cats. Can we capture this phenomenon using existing metrics? Unfortunately, we cannot, because these metrics do not provide any interpretability beyond "diversity". In this context, we propose a new evaluation protocol that measures the divergence of a set of generated images from the training set regarding the distribution of attribute strengths as follows. Single-attribute Divergence (SaD) measures the divergence regarding PDFs of a single attribute. Paired-attribute Divergence (PaD) measures the divergence regarding joint PDFs of a pair of attributes. They provide which attributes the models struggle. For measuring the attribute strengths of an image, we propose Heterogeneous CLIPScore (HCS) which measures the cosine similarity between image and text vectors with heterogeneous initial points. With SaD and PaD, we reveal the following about existing generative models. ProjectedGAN generates implausible attribute relationships such as a baby with a beard even though it has competitive scores of existing metrics. Diffusion models struggle to capture diverse colors in the datasets. The larger sampling timesteps of latent diffusion model generate the more minor objects including earrings and necklaces. Stable Diffusion v1.5 better captures the attributes than v2.1. Our metrics lay a foundation for explainable evaluations of generative models.

Related papers

• Distributionally Generative Augmentation for Fair Facial Attribute Classification [69.97710556164698]
  Facial Attribute Classification (FAC) holds substantial promise in widespread applications. FAC models trained by traditional methodologies can be unfair by exhibiting accuracy inconsistencies across varied data subpopulations. This work proposes a novel, generation-based two-stage framework to train a fair FAC model on biased data without additional annotation.
  arXiv  Detail & Related papers  (2024-03-11T10:50:53Z)
• Balancing Act: Distribution-Guided Debiasing in Diffusion Models [31.38505986239798]
  Diffusion Models (DMs) have emerged as powerful generative models with unprecedented image generation capability. DMs reflect the biases present in the training datasets. We present a method for debiasing DMs without relying on additional data or model retraining.
  arXiv  Detail & Related papers  (2024-02-28T09:53:17Z)
• Training Class-Imbalanced Diffusion Model Via Overlap Optimization [55.96820607533968]
  Diffusion models trained on real-world datasets often yield inferior fidelity for tail classes. Deep generative models, including diffusion models, are biased towards classes with abundant training images. We propose a method based on contrastive learning to minimize the overlap between distributions of synthetic images for different classes.
  arXiv  Detail & Related papers  (2024-02-16T16:47:21Z)
• MAUVE Scores for Generative Models: Theory and Practice [95.86006777961182]
  We present MAUVE, a family of comparison measures between pairs of distributions such as those encountered in the generative modeling of text or images. We find that MAUVE can quantify the gaps between the distributions of human-written text and those of modern neural language models. We demonstrate in the vision domain that MAUVE can identify known properties of generated images on par with or better than existing metrics.
  arXiv  Detail & Related papers  (2022-12-30T07:37:40Z)
• Attribute Graphs Underlying Molecular Generative Models: Path to Learning with Limited Data [42.517927809224275]
  We provide an algorithm that relies on perturbation experiments on latent codes of a pre-trained generative autoencoder to uncover an attribute graph. We show that one can fit an effective graphical model that models a structural equation model between latent codes. Using a pre-trained generative autoencoder trained on a large dataset of small molecules, we demonstrate that the graphical model can be used to predict a specific property.
  arXiv  Detail & Related papers  (2022-07-14T19:20:30Z)
• On the Strong Correlation Between Model Invariance and Generalization [54.812786542023325]
  Generalization captures a model's ability to classify unseen data. Invariance measures consistency of model predictions on transformations of the data. From a dataset-centric view, we find a certain model's accuracy and invariance linearly correlated on different test sets.
  arXiv  Detail & Related papers  (2022-07-14T17:08:25Z)
• AttriMeter: An Attribute-guided Metric Interpreter for Person Re-Identification [100.3112429685558]
  Person ReID systems only provide a distance or similarity when matching two persons. We propose an Attribute-guided Metric Interpreter, named AttriMeter, to semantically and quantitatively explain the results of CNN-based ReID models.
  arXiv  Detail & Related papers  (2021-03-02T03:37:48Z)
• Probability Link Models with Symmetric Information Divergence [1.5749416770494706]
  Two general classes of link models are proposed. The first model links two survival functions and is applicable to models such as the proportional odds and change point. The second model links two cumulative probability distribution functions.
  arXiv  Detail & Related papers  (2020-08-10T19:49:51Z)
• On the Discrepancy between Density Estimation and Sequence Generation [92.70116082182076]
  log-likelihood is highly correlated with BLEU when we consider models within the same family. We observe no correlation between rankings of models across different families.
  arXiv  Detail & Related papers  (2020-02-17T20:13:35Z)

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.