University of Washington
ICLR 2018
Citation 370
Propose NNGP.
ICML 2019
Citation 129
Propose Neural Tangent Random Feature (NTRF).
ICML 2019
Citation 619
NIPS 2019
Citation 407
Citation 344
Propose Convolutional Neural Tangent Kernel(CNTK).
Propose Neural Tangent Kernel(NTK).
It propose a way to use NTK features for UCB exploration/exploitation.
NIPS 2019
Citation 75
The paper shows that NTK might not represent all learning capacity of the DNN.
- 深度学习理论之Neural Tangent Kernel第一讲：介绍和文献总结
- Understanding the Neural Tangent Kernel By Rajat's Blog
- Code for the blog rajatvd/NTK
- Ultra-Wide Deep Nets and the Neural Tangent Kernel (NTK)
- A Visual Exploration of Gaussian Processes
- thegregyang/NTK4A
- Some Intuition on the Neural Tangent Kernel
- NTK推导的详细过程
- CMU ML Blog: Ultra-Wide Deep Nets and the Neural Tangent Kernel (NTK)
- Some Intuition on the Neural Tangent Kernel
- 直观理解Neural Tangent Kernel
- Deep Gaussian Processes
- 知乎 -【强化学习 139】PreQN
- Effortless optimization through gradient flows
- Chang Liu - Gradient Flow
- When are Neural Networks more powerful than Neural Tangent Kernels?
- Neural networks are fundamentally Bayesian
- 知乎 - Neural Contextual Bandits with UCB-based Explor...