MEMT The Maximum Entropy Modeling Toolkit (MEMT) supports parameter estimation and prediction for maximum entropy models in discrete domains. The maximum entropy framework provides a constructive method for obtaining the unique conditional distribution p*(y|x) that satisfies a set of linear constraints and maximizes the conditional entropy H(p|f) with respect to the empirical distribution f(x). The maximum entropy distribution p*(y|x) also has a unique parametric representation in the class of exponential models, as m(y|x) = r(y|x) / Z(x) where the numerator r(y|x) is a product of exponential weights r(y|x) = prod_i alpha_i^g_i(x,y) for alpha_i = exp(lambda_i), and the denominator Z(x) = sum_y r(y|x) is required to satisfy the axioms of probability.

Current Version:   1.5 License Type:   Free

Home Site: ftp://ftp.cs.princeton.edu/pub/packages/memt/ Source Code Availability: Yes Available Binary Packages: Debian Package: No RedHat RPM Package: No Other Packages: Yes (memt-1.5.i586.tar.gz) Targeted Platforms: DEC Alpha OSF/1, HP PA-RISC 9000 HPUX, Linux/Intel, SGI IRIX, Sun SunOS Software/Hardware Requirements: None

Other Links: http://www.cs.princeton.edu/~ristad/papers/memt.html (Manual) Mailing Lists/USENET News Groups: None User Comments: None See A Screen Shot? (Not Yet)

  SAL Home   |   Numerical Analysis   |   Optimization