II'm a graduate of the University of Artificial Intelligence. I clearly remember painfully choosing between programming and mathematics when I was a student back in the 90s. At that time, programming was used to achieve known results, so it didn't hold any fascination for me, so I chose mathematics. After a while, I found out there is machine learning where you can get completely unpredictable results. I got interested and took a two-month course first, and then a full six-month course.
I enjoyed studying in a stimulating environment, and participating in all kinds of different classes and seminars, as well as doing assignments on Kaggle, where you can compete with other students.
My project was about recognizing emotions. I got my inspiration from a Microsoft project, where I saw a neural network do wonders, like telling from the photo some basic emotions a person is experiencing at the moment. I set up my mind to do something similar. Of course, you need a much bigger database to make this all work, but I'm ready to do my best to succeed.
Now, I'm studying Bayesian methods, which were used in machine learning not long before neural networks gained their popularity. In my opinion, they are a bit underestimated, and after a while both Bayesian methods and neural networks will be combined together to make predictions more accurate.