Secure Machine Learning

Secure machine learning (SML) focuses on developing machine learning techniques that protect sensitive data and model intellectual property during training and inference. Current research emphasizes methods like homomorphic encryption, secure multi-party computation, and techniques to mitigate adversarial attacks (e.g., adversarial training, robust model updates) across various model architectures, including neural networks and support vector machines. The field's significance lies in enabling the responsible use of machine learning in privacy-sensitive applications, such as healthcare and finance, while also addressing concerns about model extraction and data poisoning.