Probability Value Library
Report ID: TR-472-94Author: Yianilos, Peter N. / Ristad, Eric Sven
Date: 1994-10-00
Pages: 41
Download Formats: |Postscript|
Abstract:
Hidden Markov models (HMMs) and other time series models assign probabilities to long sequences of events. Avoiding underflow is arguably the central difficulty in calculating the probability of long event sequences. This technical report presents an elegant and efficient C library for representing and manipulating probability values without underflow. Use of the library results in simpler code whose execution time compares favorably with traditional numerical methods. Our abstraction leads to a natural extension of the IEEE floating point standard, to better support statistical computation.