LWI-SVD: Low-rank, Windowed, Incremental Singular Value Decompositions on Time-Evolving Data Sets thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

LWI-SVD: Low-rank, Windowed, Incremental Singular Value Decompositions on Time-Evolving Data Sets

Published on Oct 07, 20141978 Views

Singular Value Decomposition (SVD) is computationally costly and therefore a naive implementation does not scale to the needs of scenarios where data evolves continuously. While there are various on-l

Related categories

Chapter list

LWI-SVD: Low-rank, Windowed, Incremental Singular Value Decomposition on Time-Evolving Data Sets00:00
Introduction - 100:19
Introduction - 200:36
Singular Value Decomposition (SVD)00:37
Problem Definition01:23
Incremental SVD02:05
Motivation02:49
Key Challenges - 103:39
Key Challenges - 203:54
Main Technical Contributions - 104:10
Main Technical Contributions - 204:35
Proposed Algorithms04:47
Basic Incremental SVD [Brand06]04:51
LWI-SVD - 105:54
LWI-SVD - 206:35
LWI-SVD - 307:17
Reconstruction Errors - 108:03
Reconstruction Errors - 208:34
LWI-SVD With Restarts09:16
Experimental Evaluation11:12
Experimental Setup11:15
LWI2-SVD versus SPIRIT11:39
Accuracy and Running Time - 112:24
Accuracy and Running Time - 212:44
Conclusion - 113:08
Conclusion - 213:10
Any questions?13:41