[导读] 人工智能和机器学习领域的国际顶级会议 NeurIPS 2019 公布了接受论文, 有效提交论文 6743 篇论文, 总共有 1428 接受论文, 21.1% 接受率, 包括 36 篇 Oral,164 篇 Spotlights.
NeurIPS 是人工智能和机器学习领域的国际顶级会议, 由 NIPS 基金会负责运营. 该会议全称为神经信息处理系统大会 (Conference and Workshop on Neural Information Processing Systems,NIPS), 自 1987 年开始, 每年的 12 月份, 来自世界各地的从事 AI 和 ML 相关的专家学者和从业人士汇聚一堂. 受其名称歧义带来的压力 (部分原因是其首字母缩写具有「暧昧的内涵」, 带有性别歧视的意义),2018 年的会议名称改为 NeurIPS .
NeurIPS 2019 将在 12 月 8 号加拿大温哥华会议中心举行.
NeurIPS 2019 接受论文推荐
1. Facebook 提出跨语言预训练模型 XLM,Cross-lingual Language Model Pretraining
https://arxiv.org/pdf/1901.07291.pdf
2. 超图卷积神经网络, HyperGCN: A New Method For Training Graph Convolutional Networks on Hypergraphs
https://arxiv.org/abs/1809.02589
3. 四元知识图谱嵌入, Quaternion Knowledge Graph Embeddings
https://arxiv.org/pdf/1904.10281.pdf
4. 理解医学图像中的迁移学习, Transfusion: Understanding Transfer Learning for Medical Imaging
https://arxiv.org/pdf/1902.07208.pdf
5. 理解图神经网络的表示能力, Understanding the Representation Power of Graph Neural Networks in Learning Graph Topology
- https://arxiv.org/abs/1907.05008
- 6. Visualizing the PHATE of Neural Networks,
- https://arxiv.org/abs/1908.02831
7. 多模态元学习, Toward Multimodal Model-Agnostic Meta-Learning
- https://arxiv.org/pdf/1812.07172.pdf
- 8. A Graph Theoretic Framework of Recomputation Algorithms for Memory-Efficient Backpropagation
- https://arxiv.org/abs/1905.11722
- 9. RUBi: Reducing Unimodal Biases in Visual Question Answering Paper: http://arxiv.org/abs/1906.10169Code: http://github.com/cdancette/rubi.bootstrap.pytorch
10. 理解图神经网络中的注意力与泛化机制, Understanding Attention and Generalization in Graph Neural Networks
https://arxiv.org/pdf/1905.02850.pdf
来源: https://www.qcloud.com/developer/article/1499404