0.25
0.5
0.75
1.25
1.5
1.75
2
Parameter Learning for Bayesian Networks with Strict Qualitative Influences
Published on Oct 08, 20073407 Views
We propose a new method for learning the parameters of a Bayesian network with qualitative influences. The proposed method aims to remove unwanted (context-specific) independencies that are created by
Chapter list
Parameter Learning for Bayesian Networks with Strict Qualitative Influences00:00
Motivation00:46
Qualitative Influences pt 102:19
Qualitative Influences pt 203:38
OCML Estimates Create Independencies05:49
Specifying Minimum Margins08:50
Our Proposal09:51
How This (Sometimes) Solves the Problem11:00
What If an LR Coefficient Has the Wrong Sign?12:18
How to Determine the Value?13:17
Results on Artificial Data13:56
Results on Real Data15:26
How Many Independencies Could We Fix?16:23
Conclusions16:53