University/Organization
Department of Computer Science
University of Louisiana at Monroe
Monroe, Louisiana
Department of Computer Science
University of Louisiana at Monroe
Monroe, Louisiana
Title
A Hybrid Solution for Mitigating Adversarial Attacks on Machine Learning Models
Synopsis
Machine learning models are used in many areas, such as recognizing faces, self-driving cars, medical diagnosis, fraud detection, and personalized recommendations. However, as machine learning becomes more popular, it also becomes more vulnerable to attacks. This study thoroughly reviews various attacks, the implications on machine learning models, and existing mitigation techniques.