Published Jul 31, 2021

Ishan Misra: Self-Supervised Deep Learning in Computer Vision | Lex Fridman Podcast #206

Lex Fridman and Ishan Misra explore the transformative power of self-supervised learning in computer vision, alongside the challenges of AI in autonomous systems, the comparative intricacies of vision and language processing, and the role of contrastive and active learning in optimizing AI training.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

Yann LeCun: Dark Matter of Intelligence and Self-Supervised Learning | Lex Fridman Podcast #258
Answers 383 questions
Yann LeCun: Deep Learning, ConvNets, and Self-Supervised Learning | Lex Fridman Podcast #36
Answers 383 questions
Jitendra Malik: Computer Vision | Lex Fridman Podcast #110
Answers 383 questions
Ilya Sutskever: Deep Learning | Lex Fridman Podcast #94
Answers 383 questions
Sergey Levine: Robotics and Machine Learning | Lex Fridman Podcast #108
Answers 383 questions
Ian Goodfellow: Generative Adversarial Networks (GANs) | Lex Fridman Podcast #19
Answers 383 questions
Yoshua Bengio: Deep Learning | Lex Fridman Podcast #4
Answers 383 questions
Andrej Karpathy: Tesla AI, Self-Driving, Optimus, Aliens, and AGI | Lex Fridman Podcast #333
Answers 383 questions
Dileep George: Brain-Inspired AI | Lex Fridman Podcast #115
Answers 383 questions
Oriol Vinyals: Deep Learning and Artificial General Intelligence | Lex Fridman Podcast #306
Answers 383 questions
Michael Littman: Reinforcement Learning and the Future of AI | Lex Fridman Podcast #144
Answers 383 questions
Vladimir Vapnik: Statistical Learning | Lex Fridman Podcast #5
Answers 383 questions
Andrew Ng: Deep Learning, Education, and Real-World AI | Lex Fridman Podcast #73
Answers 383 questions
Daniel Kahneman: Thinking Fast and Slow, Deep Learning, and AI | Lex Fridman Podcast #65
Answers 383 questions
Elon Musk: Tesla Autopilot | Lex Fridman Podcast #18
Answers 383 questions

Dexa/Lex Fridman

Ishan Misra: Self-Supervised Deep Learning in Computer Vision | Lex Fridman Podcast #206

Topics covered

Popular Clips

Virtual Reality Happiness

Limits of Deep Learning

Emergent Concepts in Learning

Understanding Through Similarity

Language and Learning

Learning Through Contrast

Learning vs. Reasoning

The Quest for Meaning

Autonomous Driving Timeline

Non-Contrastive Learning

Emotion in AI

Active Learning Insights

Taxonomy of Objects

Multimodal Insights

Episode Highlights

Self-Supervised Learning

AI Challenges and Future

Vision vs. Language

Active and Contrastive Learning

Related Episodes

Yann LeCun: Dark Matter of Intelligence and Self-Supervised Learning | Lex Fridman Podcast #258

Yann LeCun: Deep Learning, ConvNets, and Self-Supervised Learning | Lex Fridman Podcast #36

Jitendra Malik: Computer Vision | Lex Fridman Podcast #110

Ilya Sutskever: Deep Learning | Lex Fridman Podcast #94

Sergey Levine: Robotics and Machine Learning | Lex Fridman Podcast #108

Ian Goodfellow: Generative Adversarial Networks (GANs) | Lex Fridman Podcast #19

Yoshua Bengio: Deep Learning | Lex Fridman Podcast #4

Andrej Karpathy: Tesla AI, Self-Driving, Optimus, Aliens, and AGI | Lex Fridman Podcast #333

Dileep George: Brain-Inspired AI | Lex Fridman Podcast #115

Oriol Vinyals: Deep Learning and Artificial General Intelligence | Lex Fridman Podcast #306

Michael Littman: Reinforcement Learning and the Future of AI | Lex Fridman Podcast #144

Vladimir Vapnik: Statistical Learning | Lex Fridman Podcast #5

Andrew Ng: Deep Learning, Education, and Real-World AI | Lex Fridman Podcast #73

Daniel Kahneman: Thinking Fast and Slow, Deep Learning, and AI | Lex Fridman Podcast #65

Elon Musk: Tesla Autopilot | Lex Fridman Podcast #18

Ishan Misra: Self-Supervised Deep Learning in Computer Vision | Lex Fridman Podcast #206

Topics covered

Popular Clips

Episode Highlights

Self-Supervised LearningIshan Misra, a research scientist at FAIR, discusses the transformative potential of self-supervised learning in computer vision. He explores how this approach could redefine AI's ability to understand the visual world with minimal human intervention.

Self-Supervised Learning

AI Challenges and FutureThe conversation explores the intricate landscape of autonomous driving and the role of AI in overcoming its challenges. Ishan Misra and Lex Fridman discuss the transition to vision-based systems and the complexities of using simulations in AI training.

AI Challenges and Future

Vision vs. Language

Active and Contrastive LearningIshan Misra explores the intricacies of contrastive and active learning in machine learning. He explains how these methods enhance model training by optimizing data relationships and selection processes.

Active and Contrastive Learning

Related Episodes