Robustness in AI
Andrew discusses the challenges of ensuring machine learning models generalize across different data distributions, particularly in real-world applications. He highlights the importance of systematically studying robustness and shares insights on a rigorous risk management process inspired by safety protocols in aviation, aimed at anticipating potential failures in AI deployments.In this clip
From this podcast
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
LIVE FROM TWIMLcon! Overcoming the Barriers to Deep Learning in Production with Andrew Ng - #304
Related Questions