Bayesian networks / learning.

Bayesian networks / learning.

Is there any plan to extend the statistic package to include Bayesian networks with experience for nodes?

Thanks.

--
Nescire autem quid ante quam natus sis acciderit, id est semper esse puerum. Quid enim est aetas hominis, nisi ea memoria rerum veterum cum superiorum aetate contexitur?
5 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.

Thank you for your question. While at the moment we do not have plans to extend the statistical library with support of Bayessian networks we investigate opportunities for possible directions for development of stat functionality. Do you have any specific requirements to the algorithm you are asking about (e.g., task dimensions, particular algorithms, etc)? Also, what is the bottleneck (the most time consuming component) of your Bayessian network based application? Thanks, Andrey

Quoting - Andrey Nikolaev (Intel)

Thank you for your question. While at the moment we do not have plans to extend the statistical library with support of Bayessian networks we investigate opportunities for possible directions for development of stat functionality. Do you have any specific requirements to the algorithm you are asking about (e.g., task dimensions, particular algorithms, etc)? Also, what is the bottleneck (the most time consuming component) of your Bayessian network based application? Thanks, Andrey

Hey,

I am mostly interested in Bayesian decision theory (from a game theoretic point of view) and machine learning. So far very few Bayesian libraries do that well -- hugin being the only one really.

Mostly, I want a good library I can use. Since I am fairly impressed by your (Intel) work, I thought I would ask.

--
Nescire autem quid ante quam natus sis acciderit, id est semper esse puerum. Quid enim est aetas hominis, nisi ea memoria rerum veterum cum superiorum aetate contexitur?

Quoting - nanometrics

Hey,

I am mostly interested in Bayesian decision theory (from a game theoretic point of view) and machine learning. So far very few Bayesian libraries do that well -- hugin being the only one really.

Mostly, I want a good library I can use. Since I am fairly impressed by your (Intel) work, I thought I would ask.

Thank you for providing the additional details. Best regards, Andrey

Leave a Comment

Please sign in to add a comment. Not a member? Join today