Model Selection of Sequence Prediction Algorithms by Compression

Du Xi, Dai Zhuang
; Proceedings of The 13th International Conference on Grammatical Inference, PMLR 57:160-163, 2017.

Abstract

This paper describes estimating performance of sequence prediction algorithms and hyperparameters by compressing the training dataset itself with the probablities predicted by the trained model. With such estimation we can automate the selection and tuning process of learning algorithms. Spectral learning algorithm are experimented with.

Cite this Paper


BibTeX
@InProceedings{pmlr-v57-xi16, title = {Model Selection of Sequence Prediction Algorithms by Compression}, author = {Du Xi and Dai Zhuang}, booktitle = {Proceedings of The 13th International Conference on Grammatical Inference}, pages = {160--163}, year = {2017}, editor = {Sicco Verwer and Menno van Zaanen and Rick Smetsers}, volume = {57}, series = {Proceedings of Machine Learning Research}, address = {Delft, The Netherlands}, month = {05--07 Oct}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v57/xi16.pdf}, url = {http://proceedings.mlr.press/v57/xi16.html}, abstract = {This paper describes estimating performance of sequence prediction algorithms and hyperparameters by compressing the training dataset itself with the probablities predicted by the trained model. With such estimation we can automate the selection and tuning process of learning algorithms. Spectral learning algorithm are experimented with.} }
Endnote
%0 Conference Paper %T Model Selection of Sequence Prediction Algorithms by Compression %A Du Xi %A Dai Zhuang %B Proceedings of The 13th International Conference on Grammatical Inference %C Proceedings of Machine Learning Research %D 2017 %E Sicco Verwer %E Menno van Zaanen %E Rick Smetsers %F pmlr-v57-xi16 %I PMLR %J Proceedings of Machine Learning Research %P 160--163 %U http://proceedings.mlr.press %V 57 %W PMLR %X This paper describes estimating performance of sequence prediction algorithms and hyperparameters by compressing the training dataset itself with the probablities predicted by the trained model. With such estimation we can automate the selection and tuning process of learning algorithms. Spectral learning algorithm are experimented with.
RIS
TY - CPAPER TI - Model Selection of Sequence Prediction Algorithms by Compression AU - Du Xi AU - Dai Zhuang BT - Proceedings of The 13th International Conference on Grammatical Inference PY - 2017/01/16 DA - 2017/01/16 ED - Sicco Verwer ED - Menno van Zaanen ED - Rick Smetsers ID - pmlr-v57-xi16 PB - PMLR SP - 160 DP - PMLR EP - 163 L1 - http://proceedings.mlr.press/v57/xi16.pdf UR - http://proceedings.mlr.press/v57/xi16.html AB - This paper describes estimating performance of sequence prediction algorithms and hyperparameters by compressing the training dataset itself with the probablities predicted by the trained model. With such estimation we can automate the selection and tuning process of learning algorithms. Spectral learning algorithm are experimented with. ER -
APA
Xi, D. & Zhuang, D.. (2017). Model Selection of Sequence Prediction Algorithms by Compression. Proceedings of The 13th International Conference on Grammatical Inference, in PMLR 57:160-163

Related Material