Deep Dive into Deep Learning


(T) There are many objects which look like a bicycle but are not a bicycle. A person will make the difference between all those objects and tell accurately which one is truly a bicycle or not. But for a computer that task is much more difficult. That is what Deep Learning, the hottest research in Machine Learning, aims to change. To do so, Deep Learning is getting back to the root of Artificial Intelligence by revisiting how the brain works in order to build new Machine Learning algorithms. Deep Learning algorithms are framed to learn from multiple layers of representation and learn features from unlabeled data. There are many approaches to Deep Learning in particular convolutional neural networks, deep belief networks (DBN), restricted Boltzmann machines (RBM), and stacked auto-encoders. Deep Learning problems include computer vision, image search, speech recognition, music recognition, Web search, machine translation, natural language processing, and many others.

Following are a few materials to learn more about Deep Learning from the top minds in that field:

Professor Geoffrey Hinton from the University of Toronto and Google:

Professor Yann LeCun from NYU and Facebook:

Professor LeCun’s class on Deep Learning @ NYU

Professor Andrew Ng from Stanford, Coursera and Baidu:

Other Materials

Note: The picture above is Picasso’s Nu Couche from the Stein Collection.

Copyright © 2005-2014 by Serge-Paul Carrasco. All rights reserved.
Contact Us: asvinsider at gmail dot com.