Artificial Intelligence
275 views
Adversarial Robustness
Quick Definition
AI model ability to resist adversarial perturbations
Full Definition
The ability of AI models to maintain correct predictions when facing adversarial perturbations or attacks.
Examples
adversarial training, certified defenses, robust optimization
Related Terms
adversarial-attack
ai-safety
More Artificial Intelligence Terms
Meta-Learning
Designing models that learn how to learn new tasks
AI Alignment
Ensuring AI systems act in accordance with human values
Grounding
Connecting AI outputs to verifiable real-world sources
Transfer Learning
Adapting a pre-trained model for a different related task
Anomaly Detection
Identifying unusual patterns that deviate from expected behavior
AI Ethics
Study of moral implications in AI development