Contrastive learning is a discriminative approach that aims at grouping similar samples closer and diverse samples far from each other. It it an efficient technique to train an encoder generating distinguishable and informative representations, and it may even increase the encoder's transferability. Most current applications of contrastive learning benefit only a single representation from the last layer of an encoder.In this paper, we propose a multi-level contrasitive learning approach which applies contrastive losses at different layers of an encoder to learn multiple representations from the encoder. Afterward, an ensemble can be constructed to take advantage of the multiple representations for the downstream tasks. We evaluated the proposed method on few-shot learning problems and conducted experiments using the mini-ImageNet and the tiered-ImageNet datasets. Our model achieved the new state-of-the-art results for both datasets, comparing to previous regular, ensemble, and contrastive learing (single-level) based approaches.
翻译:对比性学习是一种歧视性方法,旨在将相似的样本更接近、更多样化的样本相互区分,这是一种有效的技术,可以训练一个能产生可区分和资料性陈述的编码器,甚至可以增加编码器的可转移性。目前大多数对比性学习应用只有利于从编码器最后一层中划出一个单一的表示。在本文中,我们建议一种多层次的对比性学习方法,在编码器的不同层中应用对比性损失来从编码器中学习多个表达。随后,可以构建一个共合体,利用下游任务的多重表述。我们评估了关于少数截图学习问题的拟议方法,并利用微型图像网络和分层图像网络数据集进行了实验。我们的模式与以往的常规、多元素和对比性循环(单层)方法相比,实现了两个数据集的新状态的艺术结果。