MIT 6.S191 (2020): Neurosymbolic AI

MIT Introduction to Deep Learning 6.S191: Lecture 7
Neurosymbolic Hybrid Artificial Intelligence
Lecturer: David Cox
January 2020

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

Lecture Outline
0:00 - Introduction
1:25 - Evolution of AI
7:32 - MIT-IBM Watson AI Lab
10:10 - Why is AI today "narrow"?
19:17 - Out-of-distribution performance
21:07 - ObjectNet
23:04 - Adversarial examples
25:24 - When does deep learning struggle?
27:00 - Neural networks vs symbolic AI
28:38 - Neurosymbolic AI
34:20 - Advantages of combining symbolic AI
39:01 - CLEVERER and more
40:40 - Summary

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!!