Search found 430 matches
- Thu Jan 07, 2010 9:14 pm
- Forum: SMILE
- Topic: Guidance needed --Diagnosing faults in solar power plants?
- Replies: 3
- Views: 4033
Re: Guidance needed --Diagnosing faults in solar power plant
Steve, Thanks for your kind words. Your question is quite general to the point of me not knowing where to start. With a Bayesian network, you can model the joint p.d. over a potentially large number of variables. Once you have a model like this, you can compute the relevance of your observations (e....
- Thu Dec 10, 2009 10:10 pm
- Forum: SMILE
- Topic: Diagnosis Documentation Correction
- Replies: 1
- Views: 2986
Cost of diagnosis
Thanks! It looks like you are right. The documentation was a student project and may contain errors here and there. I have just made a correction to this section, although this was a quick fix -- I hope I haven't overlooked something. If you notice a problem in the future, please feel free to edit t...
- Thu Dec 10, 2009 5:30 pm
- Forum: SMILE
- Topic: Diagnosis: Efficient Ranking of Joint Observations
- Replies: 1
- Views: 2716
Re: Diagnosis: Efficient Ranking of Joint Observations
Bob, You have hit the reason why we don't have automatic support for this: to pre-compute all combinations of pairs, we would need a combinatorial number of computations. If you add triples, quadruples, etc., it becomes quite something :-). Your solution is what I would do in this case. The problem ...
- Fri Nov 27, 2009 8:35 am
- Forum: SMILE
- Topic: Learning Influence Diagrams from Data
- Replies: 2
- Views: 4270
Re: Learning Influence Diagrams from Data
This is a great topic for research. As far as I know, there has been work on this or related topics but all done outside of our lab, so we can't help much. Please start your search from the proceedings of the Annual Conference on Uncertainty in Artificial Intelligence, http://uai.sis.pitt.edu/. I ho...
- Fri Sep 18, 2009 6:18 am
- Forum: SMILE
- Topic: How to find joint marginal distribution?
- Replies: 4
- Views: 7181
Ah ok thanks. Not a big problem. Edit: I noticed in GeNIe there's a button called P(e). Looks like for BNs, you can set the evidence as the joint marginal probability you are interested in, and it will calculated the P(evidence). Yes, this is one way of computing individual elements of the joint p....
- Thu Sep 17, 2009 8:25 pm
- Forum: GeNIe
- Topic: Diagnostic Value
- Replies: 1
- Views: 3370
Re: Diagnostic Value
What is the meaning of the diagnostic value numbers and how they are calculated? I understand that the higher values mean greater diagnostic value, but does a specific value, for example 0.50, be interpreted as capable of reducing uncertainty by some specific amount? This is number is, roughly spea...
- Thu Sep 17, 2009 8:21 pm
- Forum: SMILE
- Topic: How to find joint marginal distribution?
- Replies: 4
- Views: 7181
Re: How to find joint marginal distribution?
When I update beliefs in an influence diagram, the values in the chance nodes are conditional marginal probability distributions (conditioned on the informational nodes and/or decision nodes). Can I obtain joint marginal distributions by calling some method? Basically instead of P(node A = state0),...
- Thu Sep 17, 2009 8:17 pm
- Forum: SMILE
- Topic: How to find joint marginal distribution?
- Replies: 4
- Views: 7181
Re: How to find joint marginal distribution?
When I update beliefs in an influence diagram, the values in the chance nodes are conditional marginal probability distributions (conditioned on the informational nodes and/or decision nodes). Can I obtain joint marginal distributions by calling some method? Basically instead of P(node A = state0),...
- Thu Sep 17, 2009 8:12 pm
- Forum: SMILE
- Topic: Modelling hysteresis in GeNie
- Replies: 2
- Views: 3391
Re: Modelling hysteresis in GeNie
Can anyone advise me how I might represent a hysteretic process in GeNie? For example, the probability that a lake will be eutrophic depends, in large part, upon its phosphorus loading (and resultant water column concentration). However, because of system feedbacks the probability is different depe...
- Fri Aug 14, 2009 6:03 am
- Forum: GeNIe
- Topic: Setting a Value as Evidence
- Replies: 12
- Views: 12664
how is the current situation with setting a value is evidence? Is it still planned to be implemented suddendly? Hopefully your graduate student is still interested in this topic :) I'm asking because it's really important for me that this feature will be implemented, if not i have to find another s...
- Tue Aug 04, 2009 1:31 pm
- Forum: GeNIe
- Topic: learning new network
- Replies: 3
- Views: 5545
Re: learning new network
Hi, I am new to GeNIe and have the following question. I have a network to start with (say for example the Alarm.xdsl found in the example folder with GeNIE) which I open in GeNIe and use it to generate data (using Network -> Generate Data File. If I now use this data file to learn the network (usi...
- Tue Aug 04, 2009 1:24 pm
- Forum: GeNIe
- Topic: Automatic calculation of expected value of decision node
- Replies: 3
- Views: 5339
Re: Automatic calculation of expected value of decision node
Does GeNIe have a way to automatically calculate the expected value of the first decision node in an influence diagram, if it is the child of an uncertainty node, or do I have to take the optimal expected values for each outcome of the uncertainty node and take their expectation by multiplying by t...
- Thu Jul 23, 2009 8:13 pm
- Forum: GeNIe
- Topic: Simplifying a learned network
- Replies: 3
- Views: 4502
- Wed Jul 22, 2009 6:29 am
- Forum: GeNIe
- Topic: Minor Bug - Documentation
- Replies: 3
- Views: 4344
Re: Minor Bug - Documentation
The link at bottom of Tutorial 2 in offline help does not go to Tutorial 3; it goes back to Tutorial 2 instead. Thanks for letting us know. We are phasing out the off-line .chm-based tutorial and moving to the Wiki-style on-line tutorial (http://genie.sis.pitt.edu/wiki/GeNIe_Documentation), so it i...
- Wed Jul 22, 2009 5:58 am
- Forum: SMILE
- Topic: Em based learning problem
- Replies: 19
- Views: 15033
Also out of curiosity i am also interested to find out if there are any plans to implement Gibbs Sampling for belief network side of things and if not is it possible to extend smile to add it. My mentor had originally recommended i use Gibbs sampling, that's the only reason i am interested to find ...