ANOFS: Automated negotiation based online feature selection method

Abstract

Feature selection is an important technique in machine learning and pattern classification. Most existing studies of feature selection are using the batch learning methods. Such methods are not appropriate for real-world applications especially when data arrive sequentially. Recently, this problem is addressed by some feature selection techniques using online learning. Despite the advantages in efficiency of online feature selection methods, they are not always accurate enough whenhandling real world data. In this paper, we address this limitation by the integration of automated negotiation process. We present a novel method based on negotiation theory for online feature selection (ANOFS) and demonstrate its application to several public datasets.

A New Convolutional Neural Network based on a Saprse Convolutional Layer for Animal Face Detection

“…This sparse method has been accurate enough to create a weight vector for binary classification. The same fraction of the selected features (10% of all dimensions and the rest of the weights were zeros) used by [16,17,18] were chosen for use. BenSaid et al…”

“…[ 16,18] proved the effectiveness of the prediction performance of this fraction on several public large-scale benchmark datasets and thereby each pooling filter summarized the sparse ANOFS-CONV map by 10% of its real size.…”

“…The gradient features could thus be suited for animal face detection in images. The Automated Negotiation-based Online Feature Selection (ANOFS) is a sparse online learning method introduced by BenSaid and Alimi [16,17,18]. The aim of this method is to select a small number of features for binary classification on small databases and thereby replace the traditional optimizers (such as SGDM and ADAM) by the ANOFS, which decreases the number of layer parameters and operations.…”