Machine learning (ML) is widespread and used by industries for mission-critical applications such as fraud detection, algorithmic trading, etc. There are an endless number of applications that benefit from ML, but it requires real-world data for training the model. This inevitably poses challenges to security and privacy and cause issues related to regulatory and legal requirements. This session will provide a deep dive into the ways ML-based voice biometrics and automatic speech recognition can be integrated using trusted execution environment and zero knowledge proof to address security and privacy without compromising regulatory and legal requirements.
Learning Objectives:
Underline regulatory challenges associated with machine-learning models.
Recognize how trusted execution environment and zero knowledge proof can be used to address some of the security and privacy challenges associated with machine learning.
Apply the ideas presented in this topic to other machine-learning use cases.
