MIT 6.S191: Taming Dataset Bias via Domain Adaptation
MIT Introduction to Deep Learning 6.S191: Lecture 10
Taming Dataset Bias via Domain Adaptation
Lecturer: Prof. Kate Saenko, MIT-IBM Watson AI Lab
January 2021
For all lectures, slides, and lab materials: http://introtodeeplearning.com
Lecture Outline
0:00 - Introduction
3:20 - When does dataset bias occur?
7:00 - Implications in the real-world
12:41 - Dealing with data bias
14:38 - Adversarial domain alignment
20:30 - Pixel space alignment
26:03 - Few-shot pixel alignment
33:56 - Moving beyond alignment
38:59 - Enforcing consistency
42:05 - Summary and 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!!