Zobrazeno 1 - 10
of 21
pro vyhledávání: '"Nick Littlestone"'
Publikováno v:
Machine Learning. 43:173-210
The problem of learning linear-discriminant concepts can be solved by various mistake-driven update procedures, including the Winnow family of algorithms and the well-known Perceptron algorithm. In this paper we define the general class of “quasi-a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::5761993cc7ab2376325ec6866dad41a9
https://doi.org/10.1023/a:1010844028087
https://doi.org/10.1023/a:1010844028087
Publikováno v:
Information and Computation. 161:85-139
In the standard on-line model the learning algorithm tries to minimizethe total number of mistakes made in a series of trials. On each trial the learner sees an instance, makes a prediction of its classification, then finds out the correct classifica
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::030c27f50e9e2ce628bb34c86a89cea8
https://doi.org/10.1006/inco.2000.2870
https://doi.org/10.1006/inco.2000.2870
Publikováno v:
STOC
We present an algorithm for the on-line learning of linear functions which is optimal to within a constant factor with respect to bounds on the sum of squared errors for a worst case sequence of trials. The bounds are logarithmic in the number of var
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ad35e14d7963abb38a3687ca71b7a312
https://doi.org/10.1007/bf01277953
https://doi.org/10.1007/bf01277953
Publikováno v:
Journal of Computer and System Sciences. 50(1):32-40
This paper addresses the problem of learning boolean functions in query and mistake-bound models in the presence of irrelevant attributes. In learning a concept, a learner may observe a great many more attributes than those that the concept depends u
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::903a1b2eb64c13d606044cd1eba75007
Autor:
Manfred K. Warmuth, Nick Littlestone
Publikováno v:
FOCS
We study the construction of prediction algorithms in a situation in which a learner faces a sequence of trials, with a prediction to be made in each, and the goal of the learner is to make few mistakes. We are interested in the case that the learner
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::198169fb0797ca052113ed54df49eb08
Publikováno v:
Information and Computation. 95:129-161
In this paper we consider several variants of Valiant's learnability model that have appeared in the literature. We give conditions under which these models are equivalent in terms of the polynomially learnable concept classes they define. These equi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fbd7fcb9bffe7fa95260faf4aecefe77
https://doi.org/10.1016/0890-5401(91)90042-z
https://doi.org/10.1016/0890-5401(91)90042-z
Publikováno v:
COLT
The problem of learning linear-discriminant concepts can be solved by various mistake-driven update procedures, including the Winnow family of algorithms and the well-known Perceptron algorithm. In this paper we define the general class of "quasi-add
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::06ab26fff63dc26a49669ce46fda3ba8
https://doi.org/10.1145/267460.267493
https://doi.org/10.1145/267460.267493
Autor:
Nick Littlestone
Publikováno v:
ICML
Using simulations, we compare several linear-threshold learning algorithms that differ greatly in the effect of superfluous attributes on their learning abilities. These include a Bayesian algorithm for conditionally independent attributes and two mi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::596a7488e0128022c072f804333a4735
https://doi.org/10.1016/b978-1-55860-377-6.50051-7
https://doi.org/10.1016/b978-1-55860-377-6.50051-7
