ICLR 2020 元学习(Meta-learning)及小样本学习(Few-shot learning)等方向上接收论文总结

ICLR 2020 将于2020年4 月 26 日于埃塞俄比亚首都亚的斯亚贝巴举行。

在最终提交的 2594 篇论文中,有 687 篇被接收,接收率为 26.5%;和之前一样,所有被接收的论文都将进行 poster 展示;23% 的接收论文将进行口头讲述报告,其中108 篇论文为 spotlight,演讲时间 4 分钟,48 篇论文将进行 10 分钟以上的 talk

对ICLR 2020接收论文中与元学习(Meta-learning)及小样本学习(few-shot learning)相关方向的论文进行总结。若有遗漏或其他错误还请指正。

元学习(Meta-learning)

  1. Learning To Balance: Bayesian Meta-learning For Imbalanced And Out-of-distribution Tasks (Talk)
  2. Meta-learning With Warped Gradient Descent (Talk)
  3. Meta-learning Without Memorization (Spotlight)
  4. Meta-learning Acquisition Functions For Transfer Learning In Bayesian Optimization (Spotlight)
  5. Automated Relational Meta-learning (Poster)
  6. Empirical Bayes Transductive Meta-learning With Synthetic Gradients (Poster)
  7. Meta-learning Curiosity Algorithms (Poster)
  8. Towards Fast Adaptation Of Neural Architectures With Meta Learning (Poster)
  9. Differentially Private Meta-learning (Poster)
  10. Meta-learning Deep Energy-based Memory Models (Poster)
  11. Varibad: A Very Good Method For Bayes-adaptive Deep Rl Via Meta-learning (Poster)
  12. ES-MAML: Simple Hessian-Free Meta Learning (Poster)
  13. Watch, Try, Learn: Meta-learning From Demonstrations And Rewards (Poster)

Meta Reinforcement Learning

  1. Meta-Q-Learning (Talk)
  2. Improving Generalization In Meta Reinforcement Learning Using Neural Objectives (Spotlight)
  3. Meta Reinforcement Learning With Autonomous Inference Of Subtask Dependencies (Poster)

小样本学习(Few-shot Learning)

  1. A Theoretical Analysis Of The Number Of Shots In Few-shot Learning (Poster)
  2. A Baseline For Few-shot Image Classification (Poster)
  3. Metapix: Few-shot Video Retargeting (Poster)
  4. Cross-domain Few-shot Classification Via Learned Feature-wise Transformation (Spotlight)
  5. Few-shot Learning On Graphs Via Super-classes Based On Graph Spectral Measures (Poster)
  6. Few-shot Text Classification With Distributional Signatures (Poster)
  7. Meta-dataset: A Dataset Of Datasets For Learning To Learn From Few Examples (Poster)

Zero-shot Learning

  1. Locality And Compositionality In Zero-shot Learning (Poster)

参考资料:

  1. https://openreview.net/group?id=ICLR.cc/2020/Conference
  2. https://github.com/shaohua0116/ICLR2020-OpenReviewData#Data
  3. https://chillee.github.io/OpenReviewExplorer/

你可能感兴趣的:(小样本学习)