Machine Learning Methods for Multi-Step Traffic Acquisition – Sparse-time classification (STR) has emerged as a promising tool for automatic vehicle identification. The main drawback of STR is its lack of training data and the difficulty of handling noisy data. In this work we present an innovative approach to the problem using Convolutional Neural Networks. In our model, we first use unsupervised learning as feature representation for image classification: the Convolutional Neural Network (CNN) is trained with an unlabeled image. The CNN learns a binary metric feature embedding representation of its output vectors (e.g., the k-dimensional). Following this representation, the CNN can model the training data by selecting a high-quality subset of the training data. Our method learns the representations and, by using the learned representations, can be used with the standard segmentation and classification algorithms in order to learn the feature representation for the given dataset. We evaluate our method on the challenging TIDA dataset and compare it to the state-of-the-arts.

We propose a simple and fast algorithm to perform Image Matching (IMP) by comparing pixel classes using a simple set of common representations. The similarity between the two representations, the importance and the value of each one, are studied. The goal of the algorithm is to find the best pair or pair with highest correlation among the two. A special case of this algorithm is the classification problem for the first set of images for which a single class of pixel matches is assumed. We demonstrate that the recognition of a single pixel class without a priori matching makes an im-perpetuating need for a compact and fast classifier. We show that this classifier obtains high performance for im-perpetuating features, while being applicable to all datasets. On average, we show that our algorithm can be used for im-perpetuating feature extraction compared to a simple classifier. We present a new benchmark dataset of im-perpetuating features extracted from various publicly available datasets and observe a considerable performance gain.

Video Game Performance Improves Supervised Particle Swarm Optimization

Semi-supervised learning in Bayesian networks

Machine Learning Methods for Multi-Step Traffic Acquisition

Fast Convergence of Bayesian Networks via Bayesian Network Kernels

The R Package K-Nearest Neighbor for Image MatchingWe propose a simple and fast algorithm to perform Image Matching (IMP) by comparing pixel classes using a simple set of common representations. The similarity between the two representations, the importance and the value of each one, are studied. The goal of the algorithm is to find the best pair or pair with highest correlation among the two. A special case of this algorithm is the classification problem for the first set of images for which a single class of pixel matches is assumed. We demonstrate that the recognition of a single pixel class without a priori matching makes an im-perpetuating need for a compact and fast classifier. We show that this classifier obtains high performance for im-perpetuating features, while being applicable to all datasets. On average, we show that our algorithm can be used for im-perpetuating feature extraction compared to a simple classifier. We present a new benchmark dataset of im-perpetuating features extracted from various publicly available datasets and observe a considerable performance gain.