Still scouting around for meaningful open data spatial -MRFs ... Just collected a few good candidates.
Showing posts with label Ising models. Show all posts
Showing posts with label Ising models. Show all posts
Emergency visits in CA counties
Let's assume that the underlying graph is the adjacency inter-county graph of the counties of CA.
Let the county label be +1 if its emergency visit rate is above the state average and -1 if otherwise.
Does this constitute an Ising model?
Approach: p-value driven
Let the county label be +1 if its emergency visit rate is above the state average and -1 if otherwise.
Does this constitute an Ising model?
Approach: p-value driven
p-values, hypothesis testing and Ising model detection, Real world county datasets
A lot of researchers seem to be using Ising priors on the context they they help achieve better error rates and hence must be informative.
Can the p-value approach be better? Yes: But requires partition function calculation or atleast tight upper bounds. TRBP looks like it will get the job done. Well, at least in most cases.
Here is a compilation of all county-datasets in US.
Can the p-value approach be better? Yes: But requires partition function calculation or atleast tight upper bounds. TRBP looks like it will get the job done. Well, at least in most cases.
Here is a compilation of all county-datasets in US.
Staregic sensor placement for sentiment detection
With the star network, we can get reasonably nice analytical results... Q: How do we extend it?
Will one sensor cut it? Probably not!
Will one sensor cut it? Probably not!
Star network + Noisy central node
Error exponent analysis for Star networks. Quite interesting results ...
Does smarter mean more dumber?
TURNS OUT, IN OUR FRAMEWORK OF ERROR EXPONENT ANALYSIS, MAP CANNOT BE BETTER THAN TRIVIAL ESTIMATION!
County crime classication in PA + Ills of approximate inference+ ergms
Well, the stability analysis of ERGMs seems to be fascinating. Quite applicable to our model(s) I'd say ..
Subscribe to:
Posts (Atom)