Semi-supervised learning (SSL) promises improved accuracy compared to training classifiers on small labeled datasets by also training on many unlabeled images. In real applications like medical imaging, unlabeled data will be collected for expediency and thus uncurated: possibly different from the labeled set in classes or features. Unfortunately, modern deep SSL often makes accuracy worse when given uncurated unlabeled data. Recent complex remedies try to detect out-of-distribution unlabeled images and then discard or downweight them. Instead, we introduce Fix-A-Step, a simpler procedure that views all uncurated unlabeled images as potentially helpful. Our first insight is that even uncurated images can yield useful augmentations of labeled data. Second, we modify gradient descent updates to prevent optimizing a multi-task SSL loss from hurting labeled-set accuracy. Fix-A-Step can repair many common deep SSL methods, improving accuracy on CIFAR benchmarks across all tested methods and levels of artificial class mismatch. On a new medical SSL benchmark called Heart2Heart, Fix-A-Step can learn from 353,500 truly uncurated ultrasound images to deliver gains that generalize across hospitals.
翻译:半监督的学习(SSL) 与培训小标签数据集分类员相比, 其准确性会提高, 与培训小标签数据集分类员相比, 还要对许多未贴标签的图像进行培训。 在像医学成像这样的真实应用中, 未贴标签的数据会为方便而收集, 因而不精确: 可能不同于分类或特征的标签设置。 不幸的是, 现代深层次的 SLS 在给标签的未贴标签的数据提供未贴标签的数据时, 其准确性往往会更差。 最近的复杂补救措施试图检测未贴标签的图像, 然后丢弃或降低其重量。 相反, 我们引入了 Fix- A- Step, 一个简单的程序, 将所有未贴标签的未贴标签的图像都看成有潜在帮助。 我们的第一个洞察力是, 即使未贴标签的图像也能产生有用的增强标签数据。 其次, 我们修改梯度基底值更新, 防止多任务 SLSL损失因伤害标签设定的准确性而优化。 修补许多常见的深层次的 SLSL 方法, 在所有测试的方法和人造级不匹配之间提高CIFAR基准的准确性基准的准确性。 在新的SLSLSLSD- A- six- A- Sup- Sup- Sup- Sup- Sup- Supylveylvelvelvelml 至lmal 至s ass prals sups 至超超超超越35的全的医院能够真正取得的图像到35 。</s>