Slow subspace learning from stationary processes thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Slow subspace learning from stationary processes

Published on Feb 25, 20073125 Views

The talk presents a method of unsupervised learning from stationary, vector-valued processes. The method selects a subspace on the basis of an objective which can be used to bound the expected classif

Related categories

Chapter list

BOUNDS FOR LINEAR MTL00:02
ingredients of linear MTL00:34
objective04:10
error bound05:18
Rademacher complexity09:37
Hölder’s inequality11:40
theorem13:28
multi-task subspace learning23:39
error bound30:15
theorem33:40
Hölder’s inequality35:30