通过学习兼容表示实现可重用的网络组件
Towards Reusable Network Components by Learning Compatible Representations
摘要：
本文提出的第一步，兼容，因此可重用的网络组件。我们不是为不同的任务单独训练网络，而是调整训练过程以产生跨任务兼容的网络组件。特别是，我们将网络分为两个组件，一个特征提取器和一个目标任务头，并提出了各种方法来实现它们之间的兼容性。我们系统地分析了这些方法的任务标准数据集上的图像分类。我们证明，我们可以生产直接兼容的组件，而无需对原始任务进行任何微调或损害精度。之后，我们展示了兼容组件在三个应用中的使用：无监督领域自适应，在不同架构的特征提取器之间转移分类器，以及提高迁移学习的计算效率。
英文摘要：
This paper proposes to make a first step towards compatible and hence reusable network components. Rather than training networks for different tasks independently, we adapt the training process to produce network components that are compatible across tasks. In particular, we split a network into two components, a features extractor and a target task head, and propose various approaches to accomplish compatibility between them. We systematically analyse these approaches on the task of image classification on standard datasets. We demonstrate that we can produce components which are directly compatible without any fine-tuning or compromising accuracy on the original tasks. Afterwards, we demonstrate the use of compatible components on three applications: Unsupervised domain adaptation, transferring classifiers across feature extractors with different architectures, and increasing the computational efficiency of transfer learning.
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。
