Researchers at the University of Wisconsin-Madison and Amherst College have recently introduced a new form of steganography in the domain of machine learning called “training set camouflage.” Their framework, outlined in a paper pre-published on arXiv, allows a machine learning agent to hide the intention and goal of a task from a third-party observer.

Source: Techxplore

Leave a comment

Your email address will not be published. Required fields are marked *

Top