Linear Probability Forecasting

Abstract : In this paper we consider two online multi-class classification problems: classification with linear models and with kernelized models. The predictions can be thought of as probability distributions. The quality of predictions is measured by the Brier loss function. We suggest two computationally efficient algorithms to work with these problems, the second algorithm is derived by considering a new class of linear prediction models. We prove theoretical guarantees on the cumulative losses of the algorithms. We kernelize one of the algorithms and prove theoretical guarantees on the loss of the kernelized version. We perform experiments and compare our algorithms with logistic regression.
Document type :
Conference papers
Complete list of metadatas

Cited literature [10 references]  Display  Hide  Download

https://hal.inria.fr/hal-01060645
Contributor : Hal Ifip <>
Submitted on : Thursday, November 16, 2017 - 3:56:04 PM
Last modification on : Sunday, December 17, 2017 - 1:11:24 AM
Long-term archiving on : Saturday, February 17, 2018 - 3:46:41 PM

File

ZhdanovK10.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Fedor Zhdanov, Yuri Kalnishkan. Linear Probability Forecasting. 6th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations (AIAI), Oct 2010, Larnaca, Cyprus. pp.4-11, ⟨10.1007/978-3-642-16239-8_4⟩. ⟨hal-01060645⟩

Share

Metrics

Record views

203

Files downloads

94