Machine Learning

19 readers

1 users here now

Machine learning (ML) is a field devoted to understanding and building methods that let machines "learn" – that is, methods that leverage data to improve computer performance on some set of tasks. Machine learning algorithms build a model based on sample data, known as training data, in order to make predictions or decisions without being explicitly programmed to do so. Machine learning algorithms are used in a wide variety of applications, such as in medicine, email filtering, speech recognition, agriculture, and computer vision, where it is difficult or unfeasible to develop conventional algorithms to perform the needed tasks.

founded 1 year ago

Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models (arxiv.org)

submitted 1 year ago by nsa@kbin.social to c/machinelearning@kbin.social

1 comments fedilink hide all child comments

Abstract:

We systematically study a wide variety of image-based generative models spanning semantically-diverse 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 16 modern metrics for evaluating the overall performance, fidelity, diversity, 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 16 common metrics for 8 different encoders at https://github.com/layer6ai-labs/dgm-eval.

you are viewing a single comment's thread
view the rest of the comments

[–] nsa@kbin.social 2 points 1 year ago

It seems like for creative text generation tasks, metrics have been shown to be deficient; this even holds for the new model-based metrics. That leaves human evaluation (both intrinsic and extrinsic) as the gold standard for those types of tasks. I wonder if the results from this paper (and other future papers that look automatic CV metrics) will lead reviewers to demand more human evaluation in CV tasks like they do for certain NLP tasks.