In July 2016 there was an international joint conference on artificial intelligence held in New York City where many experts and professors meet together to sha
re their recent research and the commercial use cases. Machine learning (ML), deep learning(DL) and natural language processing (NLP) were the hot topics on the agenda. There are some interesting workshops and tutorials available on the conference official website http://ijcai-16.org/. One tutorial is “Deep learning and Continuous Representations for National Language Processing” that talks about a brief history of deep neural networks(DNN) and how is the media hype on DNN, give an example of neural models for query classification.
There could be several different forms of deep neural network such as classification task, ranking task and text generation task and could be some practical use cases:
- Language translation via Machine
- Web Search
- Image captioning
- Question Answering
- Contextual entity linking
- Ad selection
The Future of Big Data
With some guidance, you can craft a data platform that is right for your organization’s needs and gets the most return from your data capital.
It is truly difficult for the majority of people to understand those concepts and mathematics algorithm as it requires pretty strong foundation knowledge in Mathematics, NN and computes similarity. Fortunately, for someone who is going to learn and paly with machine learning and deep learning, there is a simple and practical way with Microsoft Computational Network Toolkit (CNTK). In the introduction – CNTK is a framework for describing learning machines. Although intended for neural networks, the learning machines are arbitrary in that the logic of the machine is described by a series of computational steps in a Computational Network. The nice part of the CNTK is that once a computational network has been described, all the computation required to learn the network parameters is taken care of automatically.
CNTK has been the open source software so you can download either from Codeplex or GitHub. There are plenty of documentations and tutorials to show you how to download, install on your local machine and how to do the exercise with this toolkit. In the tutorial, it starts with a simple model of logistical regression by using the network description language BrainScript.
CNTK supports for multiple platforms of Linux, Windows, Azure or Docker container. The typical way is to download the source code and build the executable package with Windows 64-bits OS. Here is the reference link that you can use to get CNTK installed on your local machine.
