Gordon Hazen wrote: Does anyone know if there is a "standard" reference out there on dynamic Bayes nets (Bayes nets that model change over time)? I have several cites that treat the topic tangentially, but I don't know whether there might be something more comprehensive. Thanks!
Many thanks to everyone for helping me out with references. For your information, I append below the responses which were not copied to the UAI listserve. Gordon From: Mark A. Paskin [EMAIL PROTECTED] To: Gordon Hazen [EMAIL PROTECTED] Subject: Re: [UAI] Dynamic Bayes nets Artificial Intelligence: A Modern Approach (Russell & Norvig, 1995) has a pretty good treatment. Also see Uri Lerner's dissertation (<a href="http://www.cs.stanford.edu/~uri"; eudora="autourl"><font color="#0000FF"><u>www.cs.stanford.edu/~uri</a></u></font>). From: Kevin Patrick Murphy [EMAIL PROTECTED] Subject: Re: [UAI] Dynamic Bayes nets I have written a chapter for Mike Jordan's forthcoming textbook on graphical models, which will probably become a standard... See <font color="#0000FF"><u><a href="http://www.ai.mit.edu/~murphyk/Papers/dbnchapter.pdf"; eudora="autourl">www.ai.mit.edu/~murphyk/Papers/dbnchapter.pdf</a> </u></font>Kevin From: [EMAIL PROTECTED] (Judy Goldsmith) Subject: Re: [UAI] Dynamic Bayes nets I would be particularly interested if you found a good introduction to DBNs for planning. Please let me know what you learn here. I have started new grad students on the 94-page paper, Decision- Theoretic Planning or somesuch, by Boutilier, Dean and Hanks. It's available from JAIR, came out in '99. Judy Goldsmith From: Fidel Reyes Ramos [EMAIL PROTECTED] To: Gordon Hazen [EMAIL PROTECTED] Subject: Re: [UAI] Dynamic Bayes nets Mr. Hazen, Search for Kevin Murphy's doctoral dissertation. His topic of research is precisely Dynamic Bayesian Networks. You can download it from his web page. Kevin Murphy has written several articles on dynamic bayesian networks applied in several fields, specially genomics, and he presents Hidden Markov Models (in the discrete case) and Kalma Filtes (in the continuous one) as special cases of dynamic bayesian networks. Just read his article entitled An introduction to graphical models. Hope this can help. From: Chris Elsaesser [EMAIL PROTECTED] Subject: Re: [UAI] Dynamic Bayes nets If you get any reply, pass them on or summarize if you get a chance. I have a system that converts plans into dynamic Bayes nets. In active development right now for use in an intelligence application. Paul Lehner, Scott Musman (no longer at MITRE), and I gave a paper at a AAAI spring symposium several years ago. I'll send you copies of it and a related journal paper (so you can cite) if you are interested. Not exactly what you need, I guess. chris From: Zhang, Yimin [EMAIL PROTECTED] Subject: RE: [UAI] Dynamic Bayes nets Hi Gordon: Kevin Murphy's Ph.D. thesis (especially the appendix) seems good for your purpose. You can search his name in Google to find the thesis. Yimin Zhang Statistical Computing Team Intel China Research Center From: Eric Horvitz [EMAIL PROTECTED] There are several pieces on this. An early article is: K. Kanazawa and T. Dean. A model for projection and action. In Proceedings of the Eleventh IJCAI. AAAI/International Joint Conferences on Artificial Intelligence, August 1989. Other work includes the efforts by Paul Dagum, Adam Galper , and myself on examining links between time series efforts and Bayesian networks: P. Dagum, A. Galper, and E. J. Horvitz. Dynamic Network Models for Forecasting, Proceedings of 1992 Conference on Uncertainty in Artificial Intelligence, Stanford, CA, July 1992. <font color="#0000FF"><u><a href="http://www-ksl.stanford.edu/KSL_Abstracts/KSL-92-51.html"; eudora="autourl">http://www-ksl.stanford.edu/KSL_Abstracts/KSL-92-51.html </a></u></font>A journal version of this can be found at: P. Dagum, P., A. Galper, E. Horvitz, A. Seiver, Uncertain reasoning and forecasting, International Journal of Forecasting 11(1):73-87, March 1995. <font color="#0000FF"><u><a href="http://research.microsoft.com/~horvitz/FORECAST.HTM"; eudora="autourl">http://research.microsoft.com/~horvitz/FORECAST.HTM </a></u></font>Koller and Boyan have some more recent work on tractable solution of dynamic Bayesian networks. Russell has done some applications work. Eric Gordon Hazen Department of Industrial Engineering and Management Sciences McCormick School of Engineering and Applied Science 2145 Sheridan Road Northwestern University Evanston IL 60208-3119 Fax 847-491-8005 Phone 847-491-5673 Web: <a href="http://www.iems.nwu.edu/~hazen/"; eudora="autourl">www.iems.nwu.edu/~hazen/</a></html>
