Informed Parsimonious Inference of Prototypical Genetic Sequences
Abstract
Informed Parsimonious inference (IP) generalizes the compatibility and weighted parsimony methods, which are most often applied in biology. IP is successfully applied to infer classes of genetic sequences that are accepted by biologists. IP falls into the categories of Bayesian and Minimum Description Length methods. The biological assumptions that underly inductive inference are expressed in the form of an encoding scheme. A local search strategy is then applied to find a theory that minimizes the number of bits that are needed to encode the observations under the encoding scheme.
Cite
Text
Milosavljevic et al. "Informed Parsimonious Inference of Prototypical Genetic Sequences." Annual Conference on Computational Learning Theory, 1989. doi:10.1016/B978-0-08-094829-4.50010-6Markdown
[Milosavljevic et al. "Informed Parsimonious Inference of Prototypical Genetic Sequences." Annual Conference on Computational Learning Theory, 1989.](https://mlanthology.org/colt/1989/milosavljevic1989colt-informed/) doi:10.1016/B978-0-08-094829-4.50010-6BibTeX
@inproceedings{milosavljevic1989colt-informed,
title = {{Informed Parsimonious Inference of Prototypical Genetic Sequences}},
author = {Milosavljevic, Aleksandar and Haussler, David and Jurka, Jerzy},
booktitle = {Annual Conference on Computational Learning Theory},
year = {1989},
pages = {102-117},
doi = {10.1016/B978-0-08-094829-4.50010-6},
url = {https://mlanthology.org/colt/1989/milosavljevic1989colt-informed/}
}