论文标题
在极端任务差异的几次转移的自我训练中
Self-training for Few-shot Transfer Across Extreme Task Differences
论文作者
论文摘要
大多数射击的学习技术都是在一个标有大型的“基本数据集”上预先训练的。在不可用于预训练（例如X射线，卫星图像）的问题域中，必须诉诸于不同的“源”问题域（例如Imagenet），这可能与所需的目标任务有很大不同。传统的少量射击和转移学习技术在源和目标任务之间存在如此极大的差异的情况下失败。在本文中，我们提出了一个简单有效的解决方案，以应对这个极端域间隙：从目标域中未标记的数据上进行源域表示。我们表明，这将目标域上的单发性能提高了2.9点，这是由来自多个域的数据集组成的挑战性BSCD-FSL基准测试。我们的代码可在https://github.com/cpphoo/startup上找到。
Most few-shot learning techniques are pre-trained on a large, labeled "base dataset". In problem domains where such large labeled datasets are not available for pre-training (e.g., X-ray, satellite images), one must resort to pre-training in a different "source" problem domain (e.g., ImageNet), which can be very different from the desired target task. Traditional few-shot and transfer learning techniques fail in the presence of such extreme differences between the source and target tasks. In this paper, we present a simple and effective solution to tackle this extreme domain gap: self-training a source domain representation on unlabeled data from the target domain. We show that this improves one-shot performance on the target domain by 2.9 points on average on the challenging BSCD-FSL benchmark consisting of datasets from multiple domains. Our code is available at https://github.com/cpphoo/STARTUP.
