AI Nanodegree. First Term review

AI Nanodegree. First Term review

It has been only one day and a half that I completed the first term of the Artificial Intelligence Nanodegree from Udacity, and I feel in heaven. I did learn a lot; But also, I know, now, that I don’t know anything. It has been a tough, strenuous, arduous and very hard path.

If you are already familiar with Nanodegree, you already know the structure they have. And if you don’t, let me tell you that learning is based on doing different projects. Projects for the first term are 4. At today’s they are:

1- Solving Sudoku with AI. In this project, you’ll build a simple AI using Constraint Propagation and Search to solve Sudoku puzzles. You’ll extend this to solve Diagonal Sudokus and implement advanced Sudoku strategies such as the Naked Twins strategy.

2.- Build a Game-Playing AI Agent to win the board game Isolation.

3.- Cargo Route Planning. Here, you’ll combine your knowledge of Logic, Planning and Search to implement a system that efficiently moves cargo from their origins to their destinations using the least number of flights. The system will use propositional logic to find a path to its goals given its start state and valid actions.

4.- Translate Sign Language to Text. In the final project, you’ll use HMMs to translate Sign Language into their English Language characters. Your AI will just use images of people signing to convert those into English automatically! You’ll train your model on an available dataset of sign language and use it to classify new images of signing. Advanced students will be able to translate full sentences as well.

I did enjoy a lot while doing most of the projects. #1 introduced me some concepts that were very new to me. I need some coding skills, but I manage to code with ease. The ideas to learn were clear, and the implementation of the solution was feasible. #3 is for me, the most realistic use of AI in business. In my opinion, although it is a very simplified version of a real problem, it gives a sneak peek about applications of AI in “real life”, which is much more of my interest than using on solving games or puzzles. #4 opened my mind to possible solutions beyond gaming and business issues. I found fascinating not only the project itself but also the video lectures suggested about learning dolphins communication. Up at that point, I saw myself being able to communicate with my cats beyond scratching their head when I feel they need some love.

There is one book that needs to be read to follow the algorithms. I found video lectures and papers were not enough if you are a neophyte of the area. The book is called Artificial Intelligence: A modern approach by Peter Norvig and Stuart Russell. You can download it following the link. The first chapters in the third edition cover the first term, and the second half of the book, I guess, will cover the different specialisations included in the second end. This is the one and only book on AI up until today. It is very technical, and some coding solutions in several programming languages can be found on the internet.

My mentor on the platform, Alan, encouraged me when I got stuck. I did not abuse on asking him how to deal with things. It was pointless as I had time to search online ways of solving problems. Luckily for me, I encounter many problems and I did suffer a lot on the very first project. So I quickly learnt that even if my coding solved the problem but was computational exhausting my RAM, I might not pass the unit test required before submitting.

I have done the Nanodegree almost on a full-time basis, and despite that, I failed on ending the first two projects before the soft deadline suggested. I did learn a lot, but it is not enough to be confident and walk alone in the field of Artificial Intelligence. Most of the coding is provided, and while filling the blanks, you are tipped on what to do. So this is just a first stage if you want to pursue a career in AI.

I feel I need to break before starting the second term. I deserve it. I put almost 159 hours to complete the term. I need to think very carefully which specialisation I choose, and for that, I need more information. So I will read the missing chapters in the book.