My home page
Biography
Research
Publications
My group
Courses
Professional activities
FAQ
Personal
Papers

Daphne Koller Publications

Update rules for parameter estimation in Bayesian networks (1997)

by E. Bauer, D. Koller, and Y. Singer


Abstract: This paper re-examines the problem of parameter estimation in Bayesian networks with missing values and hidden variables from the perspective of recent work in on-line learning [Kivinen & Warmuth, 1994]. We provide a unified framework for parameter estimation that encompasses both on-line learning, where the model is continuously adapted to new data cases as they arrive, and the more traditional batch learning, where a pre-accumulated set of samples is used in a one-time model selection process. In the batch case, our framework encompasses both the gradient projection algorithm and the EM algorithm for Bayesian networks. The framework also leads to new on-line and batch parameter update schemes, including a parameterized version of EM. We provide both empirical and theoretical results indicating that parameterized EM allows faster convergence to the maximum likelihood parameters than does standard EM.


Download Information

E. Bauer, D. Koller, and Y. Singer (1997). "Update rules for parameter estimation in Bayesian networks." Proc. Thirteenth Annual Conference on Uncertainty in AI (UAI) (pp. 3-13). pdf ps.gz

Bibtex citation

@inproceedings{Bauer+al:UAI97,
  author =       "E. Bauer and D. Koller and Y. Singer",
  booktitle =    "Proc. Thirteenth Annual Conference on Uncertainty in
                 AI (UAI)",
  title =        "Update rules for parameter estimation in Bayesian
                 networks",
  pages =        "3--13",
  year =         "1997",
}

full list
Click to go to robotics Click to go to theory Click to go to CS Stanford Click to go to Stanford's Webpage
home | biography | research | papers | my group
courses | professional activities | FAQ | personal