Skip to main content

A Spectral Method that Worked Well in the SPiCe'16 Competition

Liza, Farhana Ferdousi, Grzes, Marek (2016) A Spectral Method that Worked Well in the SPiCe'16 Competition. In: Verwer, Sicco and van Zaanen, Menno and Smetsers, Rick, eds. Journal of Machine Learning Research. Volume 57: Proceedings of The 13th International Conference on Grammatical Inference. JMLR: Workshop and Conference Proceedings , 57. pp. 143-148. Journal of Machine Learning Research

PDF - Publisher pdf
Download (177kB) Preview
[img]
Preview
Official URL
http://www.jmlr.org/proceedings/papers/v57/

Abstract

We present methods used in our submission to the Sequence Prediction ChallengE (SPiCe’16) 1 . The two methods used to solve the competition tasks were spectral learning and a count based method. Spectral learning led to better results on most of the problems.

Item Type: Conference or workshop item (Proceeding)
Uncontrolled keywords: Spectral Learning, Rank, Sequence Prediction, Hyperparameters
Subjects: Q Science > Q Science (General) > Q335 Artificial intelligence
Divisions: Faculties > Sciences > School of Computing > Computational Intelligence Group
Depositing User: Marek Grzes
Date Deposited: 16 Sep 2016 14:26 UTC
Last Modified: 29 May 2019 17:50 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/57327 (The current URI for this page, for reference purposes)
  • Depositors only (login required):

Downloads

Downloads per month over past year