COLT
1999
Springer
13 years 11 months ago
1999
Springer
We propose a new boosting algorithm. This boosting algorithm is an adaptive version of the boost by majority algorithm and combines bounded goals of the boost by majority algorith...
COLT
1999
Springer
13 years 11 months ago
1999
Springer
We study the problem of PAC-learning Boolean functions with random attribute noise under the uniform distribution. We define a noisy distance measure for function classes and sho...
COLT
1999
Springer
13 years 11 months ago
1999
Springer
The empirical error on a test set, the hold-out estimate, often is a more reliable estimate of generalization error than the observed error on the training set, the training estim...
COLT
1999
Springer
13 years 11 months ago
1999
Springer
COLT
1999
Springer
13 years 11 months ago
1999
Springer
We study the learnability of Threshold functions with bounded weights using membership queries only. We show that the class Ct of Threshold functions with positive integer weights...
ALT
1999
Springer
13 years 11 months ago
1999
Springer
We are concerned with the problem of sequential prediction using a givenhypothesis class of continuously-manyprediction strategies. An e ectiveperformance measure is the minimax re...
ALT
1999
Springer
13 years 11 months ago
1999
Springer
ALT
1999
Springer
13 years 11 months ago
1999
Springer
Abstract. Blum and Blum (1975) showed that a class B of suitable recursive approximations to the halting problem is reliably EX-learnable. These investigations are carried on by sh...
ALT
1999
Springer
13 years 11 months ago
1999
Springer
ALT
1999
Springer
13 years 11 months ago
1999
Springer
We introduce a new model for learning in the presence of noise, which we call the Nasty Noise model. This model generalizes previously considered models of learning with noise. Th...