Explore the limitations and new frontiers of deep learning in this lecture from MIT's Introduction to Deep Learning course. Delve into topics such as the expressivity of neural networks, generalization of deep models, and adversarial attacks. Discover emerging areas like structure in deep learning, Bayesian deep learning, and AutoML. Learn about deep evidential regression and its applications in uncertainty estimation. Gain insights into the current challenges and future directions of deep learning research, presented by lecturer Ava Soleimany. Access additional course materials, including slides and lab exercises, through the provided link.

Syllabus

- Introduction
- Course logistics
- Upcoming guest lectures
- Deep learning and expressivity of NNs
- Generalization of deep models
- Adversarial attacks
- Limitations summary
- Structure in deep learning
- Uncertainty & bayesian deep learning
- Deep evidential regression
- AutoML
- Conclusion