0.25
0.5
0.75
1.25
1.5
1.75
2
Recurrent Predictive Models for Sequence Segmentation
Published on Oct 08, 20073427 Views
Many sequential data sets have a segmental structure, and similar types of segments occur repeatedly. We consider sequences where the underlying phenomenon of interest is governed by a small set of mo
Chapter list
Recurrent predictive models for sequence segmentation00:00
Problem setting and example00:28
Task00:50
Problem setting01:28
In this paper02:25
Problem setting and example02:57
There seems to be a segmental structure pt 103:06
There seems to be a segmental structure pt 203:30
Recurrent models03:55
This was an easy case04:15
Recurrent models - how the data is generated04:29
Problem definition05:06
Recurrent models - how the data is generated (a)05:37
Problem definition (a)05:46
There seems to be a segmental structure pt 2 (a)06:20
Problem definition (b)06:39
In the example07:05
Related work07:12
Difficulties08:23
Basic algorithm - first step09:04
Dynamic programming09:54
After dynamic programming10:16
Basic algorithm, cont.11:01
K-median approach11:12
After dynamic programming (a)11:23
K-median approach (a)11:37
K-median - algorithms12:05
Facility location approach12:41
K-median approach (b)12:54
Iterative improvement13:05
Complexity of the algorithms14:06
Experiments14:53
Temperature prediction15:12
BIC graph15:36
Data table15:43
Aerosol particle formation16:01
Haplotype prediction pt 116:57
Haplotype prediction pt 217:25
How many segments needed?17:48
Observations18:34
How many segments needed? (a)18:45
Concluding remarks18:54
Open questions19:06