Skip to main content


Showing posts with the label Adversarial Machine Learning

Developers fighting back against ML adversaries

Adversarial Machine Learning Machine learning is a field of computer science and engineering that enables computers to learn from data. It has become increasingly popular in recent years because it allows computers to improve their performance on various tasks without being explicitly programmed.  Adversarial machine learning (AML) is a type of machine learning in which the learner is pitted against an “adversary”. Adversarial training has been used to improve the generalization and accuracy of machine learning models but is also applicable to other domains such as computer vision and natural language processing. However, this technology has also raised the risk of being exploited by malicious actors. This is why developers are fighting back against machine learning adversaries. Adversarial machine learning model evasion challenges Unfortunately, malicious actors can exploit this by providing artificially generated data that intentionally resists being learned by the machine learning