如何在元学习的任务之间分配数据？
How to distribute data across tasks for meta-learning?
摘要：
元学习模型将从以前的任务中获得的知识转移到快速学习新任务。它们在每个任务具有固定数量的数据点的基准上进行训练。这个数字通常是任意的，并且不知道它如何影响测试性能。由于标记数据是昂贵的，因此在训练任务中找到标签的最佳分配可以降低成本。给定一个固定的标签预算，我们应该使用少量的高标签任务，还是使用许多任务，每个任务都有很少的标签？我们是否应该为某些任务分配更多的标签，而为其他任务分配更少的标签？我们证明：1）如果任务是同构的，则存在统一的最优分配，由此所有任务获得相同量的数据; 2）在固定预算下，在任务数量和每个任务的数据点数量之间存在权衡，具有唯一的最优解; 3）当单独训练时，较难的任务应该以较少的任务数量为代价获得更多的数据; 4）当训练简单和困难的任务时，应该将更多的数据分配给简单的任务。有趣的是，神经科学实验表明，人类的视觉技能也能从简单的任务中更好地转移。我们在数学上证明了这些结果的混合线性回归，我们的经验表明，相同的结果适用于CIFAR-FS和mini-ImageNet上的Few-Shot图像分类。我们的研究结果为在收集元学习数据时跨任务分配标签提供了指导。
英文摘要：
Meta-learning models transfer the knowledge acquired from previous tasks to quickly learn new ones. They are trained on benchmarks with a fixed number of data points per task. This number is usually arbitrary and it is unknown how it affects performance at testing. Since labelling of data is expensive, finding the optimal allocation of labels across training tasks may reduce costs. Given a fixed budget of labels, should we use a small number of highly labelled tasks, or many tasks with few labels each? Should we allocate more labels to some tasks and less to others? We show that: 1) If tasks are homogeneous, there is a uniform optimal allocation, whereby all tasks get the same amount of data; 2) At fixed budget, there is a trade-off between number of tasks and number of data points per task, with a unique solution for the optimum; 3) When trained separately, harder task should get more data, at the cost of a smaller number of tasks; 4) When training on a mixture of easy and hard tasks, more data should be allocated to easy tasks. Interestingly, Neuroscience experiments have shown that human visual skills also transfer better from easy tasks. We prove these results mathematically on mixed linear regression, and we show empirically that the same results hold for few-shot image classification on CIFAR-FS and mini-ImageNet. Our results provide guidance for allocating labels across tasks when collecting data for meta-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。
