Sign inSign up
PAC-Bayesian Learning and Domain Adaptation thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

PAC-Bayesian Learning and Domain Adaptation

Published on Jan 16, 20132989 Views

Pascal Germain

In machine learning, Domain Adaptation (DA) arises when the distribution generating the test (target) data differs from the one generating the learning (source) data. It is well known that DA is an ha

Multi-Trade-offs in Machine Learning

Related categories

Statistical LearningSupervised Learning

Chapter list

PAC-Bayesian Learning and Domain Adaptation00:00
Outline00:07
Domain Adaptation (DA) : Problem Description - 100:34
Domain Adaptation (DA) : Problem Description - 201:30
Domain Adaptation (DA) : Problem Description - 302:22
A Classical Domain Adaptation Bound (VC-dim approach)03:18
A New Domain Adaptation Bound (PAC-Bayesian approach) - 105:33
A New Domain Adaptation Bound (PAC-Bayesian approach) - 206:58
PAC-Bayesian Learning of Linear Classi er07:25
PAC-Bayesian Domain Adaptation Learning of Linear Classi er - 108:16
PAC-Bayesian Domain Adaptation Learning of Linear Classi er - 208:21
Preliminary Experimental Results08:58