Attacking and Defending Machine Learning Applications of Public Cloud

TL;DR

This paper introduces a Security Development Lifecycle tailored for machine learning applications in public cloud environments, aiming to enhance security and reduce vulnerabilities in ML-as-a-service.

Contribution

It proposes a novel SDL framework specifically designed for ML applications in cloud services, addressing adversarial attacks and security challenges.

Findings

Reduces vulnerabilities in ML-as-a-service

Improves security during development process

Lowers development costs for secure ML applications

Abstract

Adversarial attack breaks the boundaries of traditional security defense. For adversarial attack and the characteristics of cloud services, we propose Security Development Lifecycle for Machine Learning applications, e.g., SDL for ML. The SDL for ML helps developers build more secure software by reducing the number and severity of vulnerabilities in ML-as-a-service, while reducing development cost.