论文信息 - Prediction algorithms and confidence measures based on algorithmic randomness theory

Prediction algorithms and confidence measures based on algorithmic randomness theory

This paper reviews some theoretical and experimental developments in building computable approximations of Kolmogorov's algorithmic notion of randomness. Based on these approximations a new set of machine learning algorithms have been developed that can be used not just to make predictions but also to estimate the confidence under the usual iid assumption.

Alexander Gammerman | Vladimir Vovk | A. Gammerman | V. Vovk

[1] Ming Li,et al. An Introduction to Kolmogorov Complexity and Its Applications , 1997, Texts in Computer Science.

[2] Alexander Gammerman,et al. Learning by Transduction , 1998, UAI.

[3] Alexander Gammerman,et al. Pattern Recognition and Density Estimation under the General i.i.d. Assumption , 2001, COLT/EuroCOLT.

[4] Alexander Gammerman,et al. Transduction with Confidence and Credibility , 1999, IJCAI.

[5] Alexander Gammerman,et al. Machine-Learning Applications of Algorithmic Randomness , 1999, ICML.

[6] Vladimir Vapnik,et al. Statistical learning theory , 1998 .

[7] Andrew P. Sage,et al. Uncertainty in Artificial Intelligence , 1987, IEEE Transactions on Systems, Man, and Cybernetics.

[8] Ilia Nouretdinov,et al. Ridge Regression Con dence , 2001 .

[9] Vladimir Vovk,et al. Ridge Regression Confidence Machine , 2001, International Conference on Machine Learning.

[10] Vladimir Vovk,et al. The typicalness framework: a comparison with the Bayesian approach , 2004 .

[11] G. Wahba. Improper Priors, Spline Smoothing and the Problem of Guarding Against Model Errors in Regression , 1978 .