portfolioUser's headshot

Safety

In the context of AI, safety refers to the practice of ensuring systems operate without causing unintended harm, adverse effects, or behaving in undesirable ways.

MAS Emergence Safety
MAS Emergence Safety

Formalized MAS emergence misalignment; proposed safety mitigation strategies.

RL Anomaly Detection
RL Anomaly Detection

Perspective on anomaly detection challenges and future in reinforcement learning.

PEOC OOD Detection
PEOC OOD Detection

PEOC uses policy entropy for OOD detection in deep RL.