Convolutional Neural Network for Computer Vision and Natural Language Processing

Machine learning techniques are widely used in the domain of Natural Language Processing (NLP) and Computer Vision (CV), In order to capture complex and non-linear features deeper machine learning architectures become more and more popular. A lot of the state of art performance have been reported by employing deep learning techniques. Convolutional Neural Network (CNN) is one variant of deep learning architectures which has received intense attention in recent years. CNN is inspired from the domain of biology, which tries to mimic the way of how signal are processed in human brain. CNN is type of feed forward artificial neural network which are constructed by multiple layers. Signals are passed through these layers with non-linear activation functions. Within each layer, there are a lot of independent node to process the signal in different regions or aspects. CNN has achieved great success in sentence modeling, image recognition and feature detection. In this paper, we introduce the motivation, intuition, architectures and algorithm of CNN. In particular, we discuss several recent achievements of CNN in NLP and CV.