The Qx-coder
M.J. Slattery, Joan L. Mitchell
IBM J. Res. Dev
We review the principles of Minimum Description Length and Stochastic Complexity as used in data compression and statistical modeling. Stochastic complexity is formulated as the solution to optimum universal coding problems extending Shannon's basic source coding theorem. The normalized maximized likelihood, mixture, and predictive codings are each shown to achieve the stochastic complexity to within asymptotically vanishing terms. We assess the performance of the minimum description length criterion both from the vantage point of quality of data compression and accuracy of statistical inference. Context tree modeling, density estimation, and model selection in Gaussian linear regression serve as examples. © 1998 IEEE.
M.J. Slattery, Joan L. Mitchell
IBM J. Res. Dev
Corneliu Constantinescu
SPIE Optical Engineering + Applications 2009
Renu Tewari, Richard P. King, et al.
IS&T/SPIE Electronic Imaging 1996
Michael D. Moffitt
ICCAD 2009