Step 0. Motivation.
I am pretty sure you have heard that Artificial Intelligence (AI) is involved into creation of very interesting things nowadays. It helps to fight cancer, create artwork and disrupt the world’s economy.
From the finance perspective, AI is on the hype as well. Investors are actively looking for AI projects – news about startups acquired by enterprises appear every day.
Step 1. Learn.
There are many educational resources available at the moment. If you plan to dig into AI more fundamentally, instead of briefly visiting neural networks, I would suggest a book by Russell and Norvig that covers most of the modern approach to AI – “Artificial Intelligence: A modern Approach”.
There is a pretty good course “Neural Networks for Machine Learning” on Coursera that has just started (November 28, 2016). It could be useful if you are a newbie in the area.
If you hear about neural networks nowadays most likely it is in the context of deep learning.
Deep learning, as stated on Wikipedia, is a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers, composed of multiple linear and non-linear transformations.
Simply put, deep learning is a more advanced neural network that is using a straightforward idea to make it deeper and more complex.
A more advanced approach to neural networks could be found at Udacity.
The state of computer vision is extremely strong at the moment. There are many experienced professionals in the field, great research papers, crowded open source projects. But most importantly, there is a crazy number of venture capitalists who want to fund such projects. Computer vision software has never been a hotter topic in academia and the whole tech industry.
Most of the modern computer vision is driven by deep learning. Coloring, video object recognition and labeling are performed using a deep learning approach to computer vision.
One of the top people in modern computer vision/deep learning is Andrej Karpathy, who is famous for winning computer vision competitions, creating models and evangelizing deep learning for computer vision. I highly recommend Karpathy’s Stanford course.
And one more comprehensive course from the University of Toronto.
And more learning!
Personally, I am a huge fan of a curated “awesome list“, but I am sure you can easily google even more of such lists.
In the past there were not many options for teams to experiment and run neural networks – you had to pay crazy amounts of money for Amazon server or to traditional server farms. There are many options nowadays. If you are just starting, take a look at Hetzner, it is a used hardware auction.
If you are going into production, I would suggest a top choice for AI cloud by servers.com.
Good luck with your neural network experiments and don’t forget to enjoy the process of creation – that is an essential part of your future success!
InData Labs helps tech startups and enterprises explore new ways of leveraging data, implement highly complex and innovative projects, and build breakthrough AI products, using machine learning, AI and Big Data technologies. Our core services include Data Strategy Consulting, Big Data Engineering, Data Science Consulting.