论文阅读笔记《Meta-SGD: Learning to Learn Quickly for Few-Shot Learning》
核心思想 本文是在MAML的基础上进一步探索利用元学习实现无模型限制的小样本学习算法。思路与MAML和Meta-LSTM比较接近，首先MAML是利用元学习的方式获得一个较好的初始化参数，在此基础上只需要进行少量样本的微调训练就可以得到较好的结果，这种方式实现简单，但由于只对初始化参数进行学习，模型的容量有限。Meta-LSTM则是利用LSTM网络作为外层网络对内层网络的各项优化参数（学习率、衰
相关文章
- 1. 论文阅读笔记《Deep Meta-Learning: Learning to Learn in the Concept Space》
- 2. Machine Learning & Deep Learning 论文阅读笔记
- 3. 论文阅读笔记《Learning to Compare: Relation Network for Few-Shot Learning》
- 4. Meta-SGD: Learning to learn quickly for few shot learning---翻译理解
- 5. 论文阅读笔记《Learning Embedding Adaptation for Few-Shot Learning》
- 6. 论文阅读笔记 Meta-Learning for Batch Mode Active Learning
- 7. 论文阅读笔记《Meta-Learning of Neural Architectures for Few-Shot Learning》
- 8. 论文阅读笔记《Task-Agnostic Meta-Learning for Few-shot Learning》
- 9. 论文阅读笔记 | (ECCV 2018) Learning to Navigate for Fine-grained Classification
- 10. 论文阅读笔记：Learning Attribute-to-Feature Mappings for Cold-Start Recommendations
- 更多相关文章...
- • RSS 阅读器 - RSS 教程
- • PHP 实例 - AJAX RSS 阅读器 - PHP教程
- • Tomcat学习笔记（史上最全tomcat学习笔记）
- • JDK13 GA发布：5大特性解读
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. 论文阅读笔记《Deep Meta-Learning: Learning to Learn in the Concept Space》
- 2. Machine Learning & Deep Learning 论文阅读笔记
- 3. 论文阅读笔记《Learning to Compare: Relation Network for Few-Shot Learning》
- 4. Meta-SGD: Learning to learn quickly for few shot learning---翻译理解
- 5. 论文阅读笔记《Learning Embedding Adaptation for Few-Shot Learning》
- 6. 论文阅读笔记 Meta-Learning for Batch Mode Active Learning
- 7. 论文阅读笔记《Meta-Learning of Neural Architectures for Few-Shot Learning》
- 8. 论文阅读笔记《Task-Agnostic Meta-Learning for Few-shot Learning》
- 9. 论文阅读笔记 | (ECCV 2018) Learning to Navigate for Fine-grained Classification
- 10. 论文阅读笔记：Learning Attribute-to-Feature Mappings for Cold-Start Recommendations