MIT 6.S191: Evidential Deep Learning and Uncertainty

MIT Introduction to Deep Learning 6.S191: Lecture 7
Evidential Deep Learning and Uncertainty Estimation
Lecturer: Alexander Amini
January 2021

For all lectures, slides, and lab materials: http://introtodeeplearning.com​

Lecture Outline
0:00​ – Introduction and motivation
5:00​ – Outline for lecture
5:50 – Probabilistic learning
8:33 – Discrete vs continuous target learning
14:12 – Likelihood vs confidence
17:40 – Types of uncertainty
21:15 – Aleatoric vs epistemic uncertainty
22:35 – Bayesian neural networks
28:55 – Beyond sampling for uncertainty
31:40 – Evidential deep learning
33:29 – Evidential learning for regression and classification
42:05 – Evidential model and training
45:06 – Applications of evidential learning
46:25 – Comparison of uncertainty estimation approaches
47:47 – Conclusion


Subscribe to stay up to date with new deep learning lectures at MIT, or follow us @MITDeepLearning on Twitter and Instagram to stay fully-connected!!