AI 科技评论按:去年年底,Ian Goodfellow与Nicolas Papernot(是的,就是ICLR 2017的最佳论文得主之一)合作开了一个博客叫cleverhans,主要写一些关于机器学习在安全与隐私问题的文章。一个Goodfellow、一个Papernot,此二神的称呼真是般配呢。 在第一篇博客里,他俩介绍了为什么攻击机器学习要远比防御容易得多。以下是雷锋网...
2021.https://proceedings.mlr.press/v139/lewis21a/lewis21a.pdf ^
ICLR 2017-5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Workshop Track Proceedings. ICLR 2016-4th International Conference on Learning Representations, ICLR 2016, San Juan, Puerto Rico, May 2-4, 2016, Conference Track Proceedings. ...
此文章将结合Thomas Kipf经典的GCN文章和他自己写的博客一同学习和入门图卷积网络。 目录 介绍 方法 什么是图卷积网络 图卷积的具体公式 应用及实验结果 最后 介绍 许多真实的数据集具有图的特性,比如社交网络(以用户为节点),citation(以学者为节点),而当时流行的卷积网络大都更适用于图像数据,因此,怎样将机器学习运...
[1] Thomas N. Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Tra...
[1] Thomas N. Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. In 5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings, 2017b. ...
Some of the submitted Conference Track papers that are not accepted to the conference proceedings will be invited for presentation in the Workshop Track.
中科大王杰教授团队提出局部消息补偿技术,解决采样子图边缘节点邻居缺失问题,弥补图神经网络(GNNs)子图采样方法缺少收敛性证明的空白,推动 GNNs 的可靠落地。 图神经网络(Graph Neural Networks,简称 GNNs)是处理图结构数据的最有效的机器学习模型之一,也是顶会论文的香饽饽。
[12] Huang, Wenbing, et al. "Adaptive sampling towards fast graph representation learning." Advances in neural information processing systems 31 (2018).[13] Chiang, Wei-Lin, et al. "Cluster-gcn: An efficient algorithm for training deep and large graph convolutional networks." Proceedings of ...
建立在之前 Input Convex Neural Network (ICNN) [3] (ICML 2017, Amos et al., 2017, CMU) 的基础上,本文作者提出一种新型的 Input Convex Recurrent Neural Network (ICRNN) 用于具有时间关联的动态系统建模。不同于通用的神经网络结构,输入凸的神经网络要求所有隐藏层之间的权重矩阵非负,同时加入了对输入...