Raising Health

Interpretability and the AI "Clear Box" with Surya Ganguli

Oct 10, 2023
Guest
Surya Ganguli
Surya Ganguli, a neural dynamics researcher at Stanford University, joins Vijay Pande to discuss interpretability in AI, and the potential for AI to become a 'clear box'. They explore the future of AI-augmented humans and the importance of understanding AI systems. They also touch on the cost of training AI models, meta learning, regulatory frameworks, and the need for interdisciplinary collaboration.
Ask episode
Chapters
Transcript
Episode notes
1
Introduction
00:00 • 2min
2
Interpretability in AI Systems
02:06 • 11min
3
The Cost of Training AI Models and the Role of Academia
13:28 • 2min
4
Meta Learning and Making Connections
15:05 • 8min
5
The Importance of Regulatory Frameworks and Understanding AI Systems
22:53 • 3min