en
0.25
0.5
0.75
1.25
1.5
1.75
2
Sampling Table Configurations for the Hierarchical Poisson-Dirichlet Process
Published on Nov 30, 20112943 Views
Hierarchical modeling and reasoning are fundamental in machine intelligence, and for this the two-parameter Poisson-Dirichlet Process (PDP) plays an important role. The most popular MCMC sampling algo
Related categories
Chapter list
Sampling Table Configulations for the Hierarchical Poisson-Dirichlet Process00:00
Outline - 100:16
What can We Do with the Poisson-Dirichlet Process?00:21
What is the Poisson - Dirichlet Process? - 101:20
What is the Poisson - Dirichlet Process? - 202:12
Why Poisson - Dirichlet Processes? - 103:14
Why Poisson - Dirichlet Processes? - 204:54
The Chinese Restaurant Process - 105:43
The Chinese Restaurant Process - 207:05
Outline - 208:27
HPDP in the Chinese Restaurant Process Metaphor - 108:41
HPDP in the Chinese Restaurant Process Metaphor - 213:11
HPDP in the Chinese Restaurant Process Metaphor - 313:12
HPDP Statistics used in Sampling13:28
Table Indicator Representation of the HPDP15:11
Table Counts Representation of the HPDP16:10
Table Indicator Representation of the HPDP17:11
How does our Sampler Work?17:32
Outline - 318:34
Datasets, Compared Algorithms and Evaluations18:48
Perplexities19:36
Convergence Speed20:02
Outline - 420:27
Conclusion20:29
Reference21:20
Thank You21:22