Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Question on Adjacency Matrix #9

Open
uhSuiL opened this issue Jun 4, 2024 · 2 comments
Open

Question on Adjacency Matrix #9

uhSuiL opened this issue Jun 4, 2024 · 2 comments

Comments

@uhSuiL
Copy link

uhSuiL commented Jun 4, 2024

请问论文中

  1. $A^{i}=SoftMax(ReLU(E^i_1(E^i_2)^T))$ 里的 $E^i_1 , E^i_2$ 的物理意义是什么?
  2. 根据上面的公式, $A^{i}$ 似乎不满足“邻接矩阵”的实际定义?然后,Visualization of Learned Inter-series Correlation部分对于 $A^{i}$ 的热力图展示,对角线的含义如何解释?
@uhSuiL
Copy link
Author

uhSuiL commented Jun 4, 2024

如果我没有理解错,这里的 $A^{i}$ 是不是想等价于GCN里的 $\tilde{D}^{-\frac{1}{2}} \tilde{A} \tilde{D}^{-\frac{1}{2}}$ ?

@YoZhibo
Copy link
Owner

YoZhibo commented Jun 4, 2024

您好,感谢关注。 $E_{1}^{i}$$E_{2}^{i}$ 也许用对起始、目标节点的嵌入说明会更容易理解。 $A^{i}$ 主要以加权的形式表现,如果您希望它更加binary,或许可以考虑调整softmax的温度,希望能帮助到您。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants