The Perceptron algorithm versus Winnow: linear versus logarithmic mistake bounds when few input variables are relevant
From MaRDI portal
Publication:1127362
DOI10.1016/S0004-3702(97)00039-8zbMath0904.68112MaRDI QIDQ1127362
Manfred K. Warmuth, Peter Auer, Jyrki Kivinen
Publication date: 13 August 1998
Published in: Artificial Intelligence (Search for Journal in Brave)
perceptron algorithmlinear threshold functionsmultiplicative updatesmistake boundsrelevant variables
Learning and adaptive systems in artificial intelligence (68T05) Parallel algorithms in computer science (68W10)
Related Items (8)
On approximating weighted sums with exponentially many terms ⋮ Learning DNF in time \(2^{\widetilde O(n^{1/3})}\) ⋮ Selection of relevant features and examples in machine learning ⋮ Efficient learning with virtual threshold gates ⋮ Sequential correction of linear classifiers ⋮ Agnostic learning of geometric patterns ⋮ Prototype Classification: Insights from Machine Learning ⋮ Robust logics
Cites Work
- Unnamed Item
- Unnamed Item
- Unnamed Item
- Exponentiated gradient versus gradient descent for linear predictors
- On-line learning of linear functions
- Learnability and the Vapnik-Chervonenkis dimension
- A theory of the learnable
- How to use expert advice
- On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities
This page was built for publication: The Perceptron algorithm versus Winnow: linear versus logarithmic mistake bounds when few input variables are relevant
