After a few scientific studies about artificial intelligence and scrum and other stuff for students and professor. I was thinking about to create near my master of science syllabus an own road map for me to study from zero to superhero in AI.
Here is the plan: I create a few sections to begin learning from scratch
- programming language
- mathematical knowledge
- Machine Learning Algorithm
- Machine Learning Tools
- Applying to a AI Case
Frist the programming language. You have to work very well and quick with your programming language either Python, R, Java or else other coding language. You have to be good at it — no matter which. Later more. This should be at first a little overview.
Second the mathematical knowledge: You have to know in Linear Algebra, Probability and Distribution, Statistics (you will need many statistics), Vector Calculus and Matrix Decompositions.
Third Machine Learning Algorithm: You may work with Machine Learning Classification of Supervised Learning, Unsupervised Learning and Reinforcement Learning. So the whole other sub-steps for the Machine Learning Algorithms like Logistic Regression, Support Vector Machines (SVM), Q-Learning etc… — are following in the my syllabus.
Fourth Step: Maschine Learning Tools, Frameworks or Engines.
Most of them your heard before: Googles TensorFlow, Torch (maybe you heard of pyTorch), IBM Watson, Deeplearning4j, Scikit-learn, keras and many others. You see in my case the last few onces are in Python ;-)
And the last step — apply it to a AI Case study. In my thesis I wrote about a Chatbot in higher education marketing. It was pretty cool to develop in only 8 weeks a new Marketing Chatbot write the thesis about it and work with a rapid prototyping concept. So these are further skills you may to learn to work with AI.
So what would be your first step now? Tell me in the comments and follow me :-)