Tsampouka, Petroula and Shawe-Taylor, John
Constant Rate Approximate Maximum Margin Algorithms s.n.
We present a new class of perceptron-like algorithms with margin in which the “effective” learning rate, defined as the ratio of the learning rate to the length of the weight vector, remains constant. We prove that the new algorithms converge in a finite number of steps and show that there exists a limit of the parameters involved in which convergence leads to classification with maximum margin.
||online learning, maximum margin classifiers
||Electronics & Computer Science
||26 Jan 2006
||17 Apr 2017 21:51
|Further Information:||Google Scholar|
Actions (login required)