arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.07459cs.CV
收藏

PT4AL:使用自我监督的预文本任务进行主动学习

PT4AL: Using Self-Supervised Pretext Tasks for Active Learning

Georgia Institute of Technology(佐治亚理工学院);SI Analytics;Lunit Inc(Lunit公司);Hanbat National University(韩巴国立大学)

John Seon Keun Yi, Minseok Seo, Jongchan Park, Dong-Geol Choi

摘要:

标记大量数据是昂贵的。主动学习的目标是解决这个问题,它只要求从未标记的集合中注释信息量最大的数据。我们提出了一种新的主动学习方法,利用自我监督的借口任务和一个独特的数据采样器来选择既困难又有代表性的数据。我们发现,一个简单的自我监督的借口任务,如旋转预测的损失,是密切相关的下游任务的损失。在主动学习迭代之前,在未标记的集合上训练借口任务学习器,并且未标记的数据被排序并根据它们的借口任务损失分成批次。在每一次主动学习迭代中,主任务模型被用于对一批中最不确定的数据进行采样以进行注释。我们在各种图像分类和分割基准上评估了我们的方法,并在CIFAR 10,Caltech-101,ImageNet和Cityscapes上实现了令人信服的性能。我们进一步表明,我们的方法在不平衡数据集上表现良好,并且可以有效地解决冷启动问题,其中主动学习性能受到随机采样的初始标记集的影响。

英文摘要:

Labeling a large set of data is expensive. Active learning aims to tackle this problem by asking to annotate only the most informative data from the unlabeled set. We propose a novel active learning approach that utilizes self-supervised pretext tasks and a unique data sampler to select data that are both difficult and representative. We discover that the loss of a simple self-supervised pretext task, such as rotation prediction, is closely correlated to the downstream task loss. Before the active learning iterations, the pretext task learner is trained on the unlabeled set, and the unlabeled data are sorted and split into batches by their pretext task losses. In each active learning iteration, the main task model is used to sample the most uncertain data in a batch to be annotated. We evaluate our method on various image classification and segmentation benchmarks and achieve compelling performances on CIFAR10, Caltech-101, ImageNet, and Cityscapes. We further show that our method performs well on imbalanced datasets, and can be an effective solution to the cold-start problem where active learning performance is affected by the randomly sampled initial labeled set.

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。

↑