Title: VIABLE: Fast Adaptation via Backpropagating Learned Loss Show Chinese title

Title: VIABLE：通过反向传播学习损失的快速适应

Abstract: In few-shot learning, typically, the loss function which is applied at test time is the one we are ultimately interested in minimising, such as the mean-squared-error loss for a regression problem. However, given that we have few samples at test time, we argue that the loss function that we are interested in minimising is not necessarily the loss function most suitable for computing gradients in a few-shot setting. We propose VIABLE, a generic meta-learning extension that builds on existing meta-gradient-based methods by learning a differentiable loss function, replacing the pre-defined inner-loop loss function in performing task-specific updates. We show that learning a loss function capable of leveraging relational information between samples reduces underfitting, and significantly improves performance and sample efficiency on a simple regression task. Furthermore, we show VIABLE is scalable by evaluating on the Mini-Imagenet dataset. Show Chinese abstract

Abstract: 在少样本学习中，通常在测试时应用的损失函数是我们最终希望最小化的，例如回归问题的均方误差损失。 然而，考虑到在测试时样本数量很少，我们认为我们希望最小化的损失函数不一定是在少样本设置中计算梯度最合适的损失函数。 我们提出了VIABLE，一种通用的元学习扩展方法，它通过学习一个可微的损失函数，在现有基于元梯度的方法基础上，替换任务特定更新中的预定义内循环损失函数。 我们表明，学习一个能够利用样本之间关系信息的损失函数可以减少欠拟合，并在简单的回归任务上显著提高性能和样本效率。 此外，我们通过在Mini-Imagenet数据集上的评估证明了VIABLE的可扩展性。