video thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

A Transductive Framework of Distance Metric Learning by Spectral Dimensionality Reduction

Published on Feb 4, 20255997 Views

Distance metric learning and nonlinear dimensionality reduction are two interesting and active topics in recent years. However, the connection between them is not thoroughly studied yet. In this paper

Related categories

Presentation

A Transductive Framework of Distance Metric Learning by Spectral Dimensionality Reduction01:40
Metric Learning: What does it do?06:31:01
What’s good?18:38:09
Endless Learning Cycle31:25:42
How to learn?40:46:32
Wait a minute…53:53:56
And Metric Learning?80:00:11
A Metric Learning Formulation89:31:08
Graph Transduction112:44:29
The Euclidean Assumption132:21:04
And Kernels152:58:33
Learning a Kernel162:49:33
Dimensionality Reduction184:07:47
More to give: RKHS regularization199:31:11
Moving y to the weights218:48:05
The parameter λ239:55:52
Experiments: Two Moons255:50:14
Experiments: UCI Data269:49:11
Experiments: MNIST276:37:52
Beyond Euclidean303:03:47