“Discussion on how small perturbations can fool neural networks and their application in data privacy.”

“things called adversarial examples. They try to confuse the model.”

“Discussion on how small edits to data can trick machine learning models.”