Few-shot classification consists of a training phase where a model is learned on a relatively large dataset and an adaptation phase where the learned model is adapted to previously-unseen tasks with limited labeled samples. In this paper, we empirically prove that the training algorithm and the adaptation algorithm can be completely disentangled, which allows algorithm analysis and design to be done individually for each phase. Our meta-analysis for each phase reveals several interesting insights that may help better understand key aspects of few-shot classification and connections with other fields such as visual representation learning and transfer learning. We hope the insights and research challenges revealed in this paper can inspire future work in related directions.
翻译:微小的分类包括一个培训阶段,在这个阶段,在相对庞大的数据集上学习模型,在适应阶段,学习的模型适应以前未见的任务,有有限的标签样本。在本文件中,我们从经验上证明,培训算法和适应算法可以完全分离,从而可以对每个阶段进行算法分析和设计。我们对每个阶段的元分析揭示出一些有趣的洞察力,可能有助于更好地了解微小的分类的关键方面以及与其他领域的联系,例如视觉演示学习和转移学习。我们希望本文件中揭示的洞察力和研究挑战能够激励今后在相关方向上的工作。