Adversarial Robustness

Quick Definition

AI model ability to resist adversarial perturbations

The ability of AI models to maintain correct predictions when facing adversarial perturbations or attacks.

adversarial training, certified defenses, robust optimization

adversarial-attack ai-safety

Converting spoken language into written text using AI

Designing models that learn how to learn new tasks

Ensuring AI systems act in accordance with human values

Identifying and locating objects within images

Detecting human body positions and joint locations

Mechanism allowing sequence positions to attend to all others