-
# Disclosure risk metrics planning
This issue will be used to plan updates for disclosure risk metrics in `syntheval`
## Confidential data baseline assessments
* [X] Methods for identifying …
-
https://arxiv.org/abs/1610.05820
Abstract—We quantitatively investigate how machine learning models leak information about the individual data records on which they were trained. We focus on the ba…
-
Following the README on the [membership inference page](https://github.com/tensorflow/privacy/tree/master/tensorflow_privacy/privacy/membership_inference_attack) yields an error.
I am running Tenso…
-
The membership inference attack may have some errors with image data set.
-
Let's try to characterize data pollution - i.e., has this LLM been pretrained on this corpus?
Simple task: pick a random passage - chop into half. Feed first half into LLM, ask it to complete the p…
-
In the README.md, the `documentation` link fails:
```
A library for running membership inference attacks (MIA) against machine learning models. Check out the documentation.
```
-
Hello, I've been interested in membership inference attacks against FL. And I found your code on Github. I'd appreciate that if you could give a more detailed description of your attack method, since …
-
Hello, Dr.song. I read your paper《Membership Inference Attacks Against Machine Learning Models》 the other day. I am very interested in it, but I have two questions about it. First, your attack requir…
-
This site is a very useful resource! If you allow me a suggestion to improve it even further, I would invite you to consider creating a new subsection, whose title could be "Attacks against synthetic …
-
在阅读您A Probabilistic Fluctuation based Membership Inference Attack文章和代码的时候有几点不明确的地方:
1)文中的eq(12)是什么意思,我看您用monte carlo来解释,但是在代码里无论是ddpm_loss还是ddim_loss都没有关于monte carlo的部分;
2)为什么在代码实现了,ddpm_loss直接使用dif…