arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.01377cs.CVcs.LGstat.ML
收藏

基于序贯学习的领域泛化算法

Sequential Learning for Domain Generalization

University of Surrey(萨里大学);The University of Edinburgh(爱丁堡大学);Samsung AI Centre, Cambridge(三星剑桥人工智能中心)

Da Li, Yongxin Yang, Yi-Zhe Song, Timothy Hospedales

摘要:

在本文中,我们提出了一个顺序学习框架的领域泛化(DG),训练模型的问题,是强大的域转移的设计。已经提出了具有不同激励直觉的各种DG方法,但它们通常针对领域泛化的单步的进行优化-在一组领域上进行训练并泛化到另一组领域。我们的顺序学习受到终身学习理念的启发,即积累经验意味着学习$n^{th}$比学习$1^{st}$更容易。在DG中,这意味着遇到一系列域,并在每一步训练中最大限度地提高下一个域的性能。域$n$的性能则取决于之前的$n-1$学习问题。因此,通过序列反向传播意味着不仅为下一个域优化性能,而且为所有后续域优化性能。与现有方法相比,所有这些域序列的训练为基础DG学习者提供了更多的“实践”,从而提高了真正测试域的性能。这种策略可以实例化为不同的基础DG算法,但我们专注于它的应用程序最近提出的元学习领域泛化(MLDG)。我们表明,MLDG导致一个简单的实现和快速的算法,提供一致的性能改善各种DG基准。

英文摘要:

In this paper we propose a sequential learning framework for Domain Generalization (DG), the problem of training a model that is robust to domain shift by design. Various DG approaches have been proposed with different motivating intuitions, but they typically optimize for a single step of domain generalization -- training on one set of domains and generalizing to one other. Our sequential learning is inspired by the idea lifelong learning, where accumulated experience means that learning the $n^{th}$ thing becomes easier than the $1^{st}$ thing. In DG this means encountering a sequence of domains and at each step training to maximise performance on the next domain. The performance at domain $n$ then depends on the previous $n-1$ learning problems. Thus backpropagating through the sequence means optimizing performance not just for the next domain, but all following domains. Training on all such sequences of domains provides dramatically more `practice' for a base DG learner compared to existing approaches, thus improving performance on a true testing domain. This strategy can be instantiated for different base DG algorithms, but we focus on its application to the recently proposed Meta-Learning Domain generalization (MLDG). We show that for MLDG it leads to a simple to implement and fast algorithm that provides consistent performance improvement on a variety of DG benchmarks.

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。

↑