How do I poison an SVM with manifold regularization?

Question

I'm working on Adversarial Machine Learning, and have read multiple papers on this topic, some of them are mentioned as follows:

• Poisoning Attacks on SVMs: https://arxiv.org/pdf/1206.6389.pdf
• Adversarial Label Flips on Support Vector Machines

However, I am not able to find any literature on data poisoning for SVMs using Manifold regularization. Is there anyone who has knowledge about that?

Answer 1

I think, one could generate a data set, with a random variable in each component of the data vector, add this data to the training data set, and then shuffle the combined data set.