Optimizing 0/1 Loss for Perceptrons by Random Coordinate Descent

—The 0/1 loss is an important cost function for perceptrons. Nevertheless it cannot be easily minimized by most existing perceptron learning algorithms. In this paper, we propose a family of random coordinate descent algorithms to directly minimize the 0/1 loss for perceptrons, and prove their convergence. Our algorithms are computationally efficient, and usually achieve the lowest 0/1 loss compared with other algorithms. Such advantages make them favorable for nonseparable real-world problems. Experiments show that our algorithms are especially useful for ensemble learning, and could achieve the lowest test error for many complex data sets when coupled with AdaBoost.

• Type: Conference
• Year: 2007
• Where: IJCNN
• Authors: Ling Li, Hsuan-Tien Lin
• Keywords: Coordinate Descent Algorithms, IJCNN 2007, Lowest 0/1 Loss, Perceptron Learning Algorithms