[论文解读] Learning to Support: Exploiting Structure Information in Support Sets for One-Shot Learning

Jinchao Liu, Stuart Gibson+1 more|arXiv (Cornell University)|Aug 22, 2018

Domain Adaptation and Few-Shot Learning参考文献 21被引用 4

一句话总结

该论文提出了一种用于少样本学习的新颖元学习器，通过类支持网络和竞争性注意力机制，实现动态原型选择，从而增强支持集表示。通过整合更深层的CNN特征表示和近似集成方法，该方法在mini-ImageNet和Omniglot基准上取得了最先进性能，尤其在低样本设置下表现优异。

ABSTRACT

Deep Learning shows very good performance when trained on large labeled data sets. The problem of training a deep net on a few or one sample per class requires a different learning approach which can generalize to unseen classes using only a few representatives of these classes. This problem has previously been approached by meta-learning. Here we propose a novel meta-learner which shows state-of-the-art performance on common benchmarks for one/few shot classification. Our model features three novel components: First is a feed-forward embedding that takes random class support samples (after a customary CNN embedding) and transfers them to a better class representation in terms of a classification problem. Second is a novel attention mechanism, inspired by competitive learning, which causes class representatives to compete with each other to become a temporary class prototype with respect to the query point. This mechanism allows switching between representatives depending on the position of the query point. Once a prototype is chosen for each class, the predicated label is computed using a simple attention mechanism over prototypes of all considered classes. The third feature is the ability of our meta-learner to incorporate deeper CNN embedding, enabling larger capacity. Finally, to ease the training procedure and reduce overfitting, we averages the top $t$ models (evaluated on the validation) over the optimization trajectory. We show that this approach can be viewed as an approximation to an ensemble, which saves the factor of $t$ in training and test times and the factor of of $t$ in the storage of the final model.

研究动机与目标

• 解决少样本和一次样本学习中的挑战，即由于标注数据有限，深度网络通常会发生过拟合。
• 通过从少量样本中学习更优、更具判别性的原型，改善支持集中类别的表示。
• 在元学习框架中有效利用深层CNN嵌入，而此前的方法往往未能充分利用这些特征。
• 通过高效的元学习器近似集成方法，减少过拟合并提升泛化能力。

提出的方法

• 前馈式类支持网络通过学习到的变换，将原始支持样本转换为更具代表性的类别嵌入。
• 竞争性注意力机制根据查询样本的位置，动态选择每个类别的原型，实现上下文感知的原型选择。
• 通过在所选原型上应用简单的注意力机制，计算查询样本的最终类别概率。
• 通过将深层CNN嵌入（例如9层）整合到元学习器中，支持更深层的CNN嵌入，从而增强特征表达能力。
• 通过基于验证性能选择前t个模型并进行平均，形成元学习器的近似集成，降低过拟合和存储成本。
• 采用基于episode的端到端训练方式，模拟少样本分类任务。

实验结果

研究问题

• RQ1 元学习器能否有效从少量随机采样的支持集中学习到更优的类别代表？
• RQ2 基于查询位置动态选择原型的竞争性注意力机制是否能提升少样本分类性能？
• RQ3 能否在不引发过拟合的前提下，成功将深层CNN嵌入整合到基于度量的元学习框架中？
• RQ4 元学习器的近似集成在多大程度上能减少过拟合并提升泛化能力？

主要发现

• 在5类1样本和5类5样本的mini-ImageNet基准上，所提方法优于当前最先进方法，尤其在使用更深层的9层嵌入时表现更优。
• 当使用4层嵌入时，该方法在Omniglot和mini-ImageNet上均优于Prototypical Nets和Relation Nets，准确率更优。
• 在低样本设置下，特别是1样本学习中，类支持嵌入带来了最大的性能提升，因为此时随机采样的支持样本难以准确代表其类别。
• 元学习器的近似集成方法在性能上持续优于早停策略，并显著降低过拟合，尤其在高容量模型中效果明显。
• 该方法在使用比SNAIL更浅的9层CNN（而非SNAIL的13层残差网络）时，性能仍优于SNAIL，表明其对嵌入容量的利用更为高效。