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

请问有关于HeteroNN训练中Interactive Layer的噪声问题 #5280

Closed
ATowerz opened this issue Nov 26, 2023 · 3 comments
Closed

请问有关于HeteroNN训练中Interactive Layer的噪声问题 #5280

ATowerz opened this issue Nov 26, 2023 · 3 comments

Comments

@ATowerz
Copy link

ATowerz commented Nov 26, 2023

1.请问一下这个梯度下降时的噪声有参数进行控制吗?

2.我看了一下RandomNumberGenerator文件,这个噪声的生成的随机值范围是在[-1024,1024]区间,我记得在预测的时候interactive_layer的输出值和self.acc_noise累计噪声是相关的,累计噪声的数值波动大不会使host方的预测大幅度失真而导致heteroNN效果不佳吗?

@talkingwallace
Copy link
Contributor

hi 噪声在训练&预测的时候都会抵消的,不会影响到预测结果,这个目前没有相关控制参数

@ATowerz
Copy link
Author

ATowerz commented Nov 27, 2023

hi 噪声在训练&预测的时候都会抵消的,不会影响到预测结果,这个目前没有相关控制参数

您好,我还是有点疑惑,在https://github.com/FederatedAI/FATE/blob/master/doc/federatedml_component/hetero_nn.md
,interactive layer在激活前的输出值为z = z_A + epsilon_acc * alpha_A + z_B(if exists),这部分epsilon_acc* alpha_A是没有影响的吗?然后梯度下降时A方生成的随机值范围是在[-1024,1024]区间,累加后是epsilon_acc,epsilon_acc的噪声太大不会导致预测失真吗,这里我是哪里理解错误了呢

@talkingwallace
Copy link
Contributor

Hi! https://arxiv.org/pdf/2007.06849.pdf 可以参考这篇文章的,噪声在整个训练过程中是会抵消的

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

3 participants