Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models

• URL: http://arxiv.org/abs/2306.04675v2
• Date: Mon, 30 Oct 2023 18:00:00 GMT
• Title: Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models
• Authors: George Stein, Jesse C. Cresswell, Rasa Hosseinzadeh, Yi Sui, Brendan Leigh Ross, Valentin Villecroze, Zhaoyan Liu, Anthony L. Caterini, J. Eric T. Taylor, Gabriel Loaiza-Ganem
• Abstract summary: We compare 17 modern metrics for evaluating the overall performance, fidelity, diversity, rarity, and memorization of generative models. We find that the state-of-the-art perceptual realism of diffusion models as judged by humans is not reflected in commonly reported metrics such as FID. Next, we investigate data memorization, and find that generative models do memorize training examples on simple, smaller datasets like CIFAR10, but not necessarily on more complex datasets like ImageNet.
• Score: 14.330863905963442
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We systematically study a wide variety of generative models spanning semantically-diverse image datasets to understand and improve the feature extractors and metrics used to evaluate them. Using best practices in psychophysics, we measure human perception of image realism for generated samples by conducting the largest experiment evaluating generative models to date, and find that no existing metric strongly correlates with human evaluations. Comparing to 17 modern metrics for evaluating the overall performance, fidelity, diversity, rarity, and memorization of generative models, we find that the state-of-the-art perceptual realism of diffusion models as judged by humans is not reflected in commonly reported metrics such as FID. This discrepancy is not explained by diversity in generated samples, though one cause is over-reliance on Inception-V3. We address these flaws through a study of alternative self-supervised feature extractors, find that the semantic information encoded by individual networks strongly depends on their training procedure, and show that DINOv2-ViT-L/14 allows for much richer evaluation of generative models. Next, we investigate data memorization, and find that generative models do memorize training examples on simple, smaller datasets like CIFAR10, but not necessarily on more complex datasets like ImageNet. However, our experiments show that current metrics do not properly detect memorization: none in the literature is able to separate memorization from other phenomena such as underfitting or mode shrinkage. To facilitate further development of generative models and their evaluation we release all generated image datasets, human evaluation data, and a modular library to compute 17 common metrics for 9 different encoders at https://github.com/layer6ai-labs/dgm-eval.

Related papers

• Self-Consuming Generative Models with Curated Data Provably Optimize Human Preferences [20.629333587044012]
  We study the impact of data curation on iterated retraining of generative models. We prove that, if the data is curated according to a reward model, the expected reward of the iterative retraining procedure is maximized.
  arXiv  Detail & Related papers  (2024-06-12T21:28:28Z)
• Towards Theoretical Understandings of Self-Consuming Generative Models [56.84592466204185]
  This paper tackles the emerging challenge of training generative models within a self-consuming loop. We construct a theoretical framework to rigorously evaluate how this training procedure impacts the data distributions learned by future models. We present results for kernel density estimation, delivering nuanced insights such as the impact of mixed data training on error propagation.
  arXiv  Detail & Related papers  (2024-02-19T02:08:09Z)
• Learning Defect Prediction from Unrealistic Data [57.53586547895278]
  Pretrained models of code have become popular choices for code understanding and generation tasks. Such models tend to be large and require commensurate volumes of training data. It has become popular to train models with far larger but less realistic datasets, such as functions with artificially injected bugs. Models trained on such data tend to only perform well on similar data, while underperforming on real world programs.
  arXiv  Detail & Related papers  (2023-11-02T01:51:43Z)
• Learning to Jump: Thinning and Thickening Latent Counts for Generative Modeling [69.60713300418467]
  Learning to jump is a general recipe for generative modeling of various types of data. We demonstrate when learning to jump is expected to perform comparably to learning to denoise, and when it is expected to perform better.
  arXiv  Detail & Related papers  (2023-05-28T05:38:28Z)
• Feature Likelihood Divergence: Evaluating the Generalization of Generative Models Using Samples [25.657798631897908]
  Feature Likelihood Divergence provides a comprehensive trichotomic evaluation of generative models. We empirically demonstrate the ability of FLD to identify overfitting problem cases, even when previously proposed metrics fail.
  arXiv  Detail & Related papers  (2023-02-09T04:57:27Z)
• A Study on the Evaluation of Generative Models [19.18642459565609]
  Implicit generative models, which do not return likelihood values, have become prevalent in recent years. In this work, we study the evaluation metrics of generative models by generating a high-quality synthetic dataset. Our study shows that while FID and IS do correlate to several f-divergences, their ranking of close models can vary considerably.
  arXiv  Detail & Related papers  (2022-06-22T09:27:31Z)
• How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models [95.8037674226622]
  We introduce a 3-dimensional evaluation metric that characterizes the fidelity, diversity and generalization performance of any generative model in a domain-agnostic fashion. Our metric unifies statistical divergence measures with precision-recall analysis, enabling sample- and distribution-level diagnoses of model fidelity and diversity.
  arXiv  Detail & Related papers  (2021-02-17T18:25:30Z)
• Evaluating the Interpretability of Generative Models by Interactive Reconstruction [30.441247705313575]
  We introduce a task to quantify the human-interpretability of generative model representations. We find performance on this task much more reliably differentiates entangled and disentangled models than baseline approaches.
  arXiv  Detail & Related papers  (2021-02-02T02:38:14Z)
• Flow-based Generative Models for Learning Manifold to Manifold Mappings [39.60406116984869]
  We introduce three kinds of invertible layers for manifold-valued data, which are analogous to their functionality in flow-based generative models. We show promising results where we can reliably and accurately reconstruct brain images of a field of orientation distribution functions.
  arXiv  Detail & Related papers  (2020-12-18T02:19:18Z)
• Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modeling [55.28436972267793]
  Current autoencoder-based disentangled representation learning methods achieve disentanglement by penalizing the ( aggregate) posterior to encourage statistical independence of the latent factors. We present a novel multi-stage modeling approach where the disentangled factors are first learned using a penalty-based disentangled representation learning method. Then, the low-quality reconstruction is improved with another deep generative model that is trained to model the missing correlated latent variables.
  arXiv  Detail & Related papers  (2020-10-25T18:51:15Z)
• Automatic Recall Machines: Internal Replay, Continual Learning and the Brain [104.38824285741248]
  Replay in neural networks involves training on sequential data with memorized samples, which counteracts forgetting of previous behavior caused by non-stationarity. We present a method where these auxiliary samples are generated on the fly, given only the model that is being trained for the assessed objective. Instead the implicit memory of learned samples within the assessed model itself is exploited.
  arXiv  Detail & Related papers  (2020-06-22T15:07:06Z)

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.