Explore the limitations and new frontiers of deep learning in this comprehensive lecture from MIT's Introduction to Deep Learning course. Delve into topics such as the expressivity and generalization of neural networks, failure modes, uncertainty in deep learning, and adversarial attacks. Examine the challenges of algorithmic bias and discover emerging areas like learning on graphs and 3D point clouds. Gain insights into Automated Machine Learning (AutoML) and its potential impact on the field. Enhance your understanding of deep learning's current limitations and exciting future directions through this in-depth presentation by lecturer Ava Soleimany.
Overview
Syllabus
​ - Introduction
​ - Course logistics
​ - Upcoming hot topics and guest lectures
​ - Deep learning and expressivity of NNs
​ - Generalization of deep models
- Neural network failure modes
- Uncertainty in deep learning
​ - Adversarial attacks
- Algorithmic bias
​ - Limitations summary
​ - Structure in DL
- Learning on graphs
- Learning on 3D pointclouds
​ - Automated Machine Learning AutoML
- Conclusion
Taught by
https://www.youtube.com/@AAmini/videos
