ML Anthology

CarpeDiem: An Algorithm for the Fast Evaluation of SSL Classifiers

Roberto Esposito, Daniele Paolo Radicioni
ICML 2007 pp. 257-264
doi:10.1145/1273496.1273529 /icml/2007/esposito2007icml-carpediem/

Abstract

In this paper we present a novel algorithm, CarpeDiem. It significantly improves on the time complexity of Viterbi algorithm, preserving the optimality of the result. This fact has consequences on Machine Learning systems that use Viterbi algorithm during learning or classification. We show how the algorithm applies to the Supervised Sequential Learning task and, in particular, to the HMPerceptron algorithm. We illustrate CarpeDiem in full details, and provide experimental results that support the proposed approach.

PDF ICML Semantic Scholar

Cite

Text

Esposito and Radicioni. "CarpeDiem: An Algorithm for the Fast Evaluation of SSL Classifiers." International Conference on Machine Learning, 2007. doi:10.1145/1273496.1273529

Markdown

[Esposito and Radicioni. "CarpeDiem: An Algorithm for the Fast Evaluation of SSL Classifiers." International Conference on Machine Learning, 2007.](https://mlanthology.org/icml/2007/esposito2007icml-carpediem/) doi:10.1145/1273496.1273529

BibTeX

@inproceedings{esposito2007icml-carpediem,
  title     = {{CarpeDiem: An Algorithm for the Fast Evaluation of SSL Classifiers}},
  author    = {Esposito, Roberto and Radicioni, Daniele Paolo},
  booktitle = {International Conference on Machine Learning},
  year      = {2007},
  pages     = {257-264},
  doi       = {10.1145/1273496.1273529},
  url       = {https://mlanthology.org/icml/2007/esposito2007icml-carpediem/}
}