arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2303.07084cs.CV
收藏

表示学习的挑战:深度视觉模型的准确性提高并不意味着感知相似性的预测更好

The challenge of representation learning: Improved accuracy in deep vision models does not come with better predictions of perceptual similarity

Humboldt-Universität zu Berlin(柏林洪堡大学);University of Milano-Bicocca(米兰比可卡大学)

Fritz Günther, Marco Marelli, Marco Alessandro Petilli

摘要:

在过去的几年里,计算机视觉深度学习模型的进步已经导致其图像分类准确性的显着提高。然而,在训练任务中具有更高准确度的模型不一定能开发出更好的图像表示,从而使它们在其他未训练的任务中也能表现得更好。为了研究突出的高性能计算机视觉模型的表示学习能力,我们研究了它们从大规模行为数据集中捕获感知相似性的各种指数的能力。我们发现,更高的图像分类准确率与这些数据集的更好性能无关,事实上,我们观察到自GoogLeNet(2015年发布)和VGG-M(2014年发布)以来,性能没有任何改善。我们推测,更准确的分类可能是由于高度相似类之间的超工程实现了非常细粒度的区分,这并不激励模型捕捉整体感知相似性。

英文摘要:

Over the last years, advancements in deep learning models for computer vision have led to a dramatic improvement in their image classification accuracy. However, models with a higher accuracy in the task they were trained on do not necessarily develop better image representations that allow them to also perform better in other tasks they were not trained on. In order to investigate the representation learning capabilities of prominent high-performing computer vision models, we investigated how well they capture various indices of perceptual similarity from large-scale behavioral datasets. We find that higher image classification accuracy rates are not associated with a better performance on these datasets, and in fact we observe no improvement in performance since GoogLeNet (released 2015) and VGG-M (released 2014). We speculate that more accurate classification may result from hyper-engineering towards very fine-grained distinctions between highly similar classes, which does not incentivize the models to capture overall perceptual similarities.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑