![](https://img-blog.csdnimg.cn/20201014180756925.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
Few Shot Learning
文章平均质量分 50
Rhythm07
这个作者很懒,什么都没留下…
展开
-
【论文笔记 FSL 8】A Baseline for Few-Shot Image Classification(ICLR 2020)
【论文笔记 FSL 8】A Baseline for Few-Shot Image Classification(ICLR 2020)下载地址原创 2021-04-22 09:02:41 · 746 阅读 · 0 评论 -
【论文笔记 FSL 7】Free Lunch for Few-Shot Learning: Distribution Calibration(ICLR 2021)
【论文笔记 FSL 6】Instance Credibility Inference for Few-Shot Learning(CVPR2020)[下载地址](https : //openreview.net/forum?id=JWOiYxMG92s) | 论文源码原创 2021-02-03 18:57:50 · 1597 阅读 · 0 评论 -
【论文笔记 FSL 6】Instance Credibility Inference for Few-Shot Learning(CVPR2020)
【论文笔记 FSL 6】Instance Credibility Inference for Few-Shot Learning(CVPR2020)AbstractMethodExperiments1 Results下载地址 | 论文源码Abstract提出一种基于数据增强的小样本学习算法(ICI)。首先ICI使用有标签的样本训练一个线性分类器,预测测试样本的类别,得到伪标签,通过统计的方法将置信度高的测试数据加入到训练集中,达到数据增强的目的。逐渐迭代直到分类器的性能最优。Me原创 2020-12-05 15:21:57 · 385 阅读 · 0 评论 -
【论文笔记 FSL 5】DPGN: Distribution Propagation Graph Network for Few-shot Learning(CVPR2020)
论文笔记模板AbstractContributionsMethod1 Learning embedding model through classification2 Base learnerExperiments1 SetupArchitectures下载地址 | 论文源码Abstract展示了一个简单的基准:在元训练数据集上学习一个监督或自监督表示,然后在这个表示上训练一个线性分类器。通过使用自蒸馏(self-distillation)可以达到额外的促进作用。Contrib原创 2020-12-03 16:14:16 · 526 阅读 · 0 评论 -
【论文笔记 FSL 4】DeepEMD: Few-Shot Image Classification with Differentiable Earth Mover’s Dis(CVPR2020)
【论文笔记 FSL 4】DeepEMD: Few-Shot Image Classification with Differentiable Earth Mover’s Dista(CVPR2020)AbstractContributionsMethod1 Learning embedding model through classification2 Base learnerExperiments1 SetupArchitectures下载地址 | 论文源码Abstract展示原创 2020-12-03 13:24:51 · 1702 阅读 · 12 评论 -
【论文笔记 FSL 3】Distribution Consistency Based Covariance Metric Networks for FSL(AAAI2019)
【论文笔记 FSL 3】Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?(CVPR2020)AbstractContributionsMethod1 Learning embedding model through classification2 Base learner3 Sequential self-distillationExperiments1 Setup2 ResultArchitecture原创 2020-10-22 09:25:10 · 1130 阅读 · 1 评论 -
【论文笔记 FSL 1】Adaptive Subspaces for Few-Shot Learning(CVPR2020)
【论文笔记-FSL-1】Adaptive Subspaces for Few-Shot Learning原创 2020-08-31 20:46:12 · 490 阅读 · 0 评论 -
【论文笔记 FSL 2】Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?
【论文笔记-FSL-1】Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?原创 2020-08-31 14:06:45 · 2047 阅读 · 2 评论