Tips And Links on How To Learn Machine Learning!
Hello to everyone!
It’s no secret that interest in machine learning and artificial intelligence grows at best exponentially. In the meantime, my Google Drive has become a huge dumpster of papers, and the bookmarks in Google Chrome have turned into a list, the length of which tends to infinity every day.
Thus, in order to simplify the life of yourself and you, decided to structure the information and give a lot of links to interesting resources that we studied and that we recommend studying to you if you are only at the beginning of the path.
We see the path for the development of a beginner like this:
Try to start with a small start, if you do not have 6 years of VMK expertise on forecasting methods, do not immediately download the archive of lectures by E. Sokolov or K. Vorontsov, perhaps articles on Medium will be more optimal for you.
- Difficulties can also arise with understanding algorithms if you are not well versed in probability theory, optimization theory and statistics, so we advise you to stock up on math lectures.
Further, having familiarized with the theory it will be easier to apply knowledge in solving problems. Next, we will give you a list of interesting resources that we went through. And, of course, we wish you success in your journey.
Newbies:
Lifehack for quick selection of models from the team of SmartSpate:
- Data Science Glossary
- Crash-Course on basic articles on deep learning on Medium
- TensorFlow Tutorial
- Python vs. R – differences
- Excellent ML CheatSheet
- Arithmetic of Convolutional Neural Networks from Theano Team
- Machine Learning Basics
Good explanations of how ROC-AUC works:
Advance:
- Selection and assessment of models – the basis
- A book on the natural language toolkit
- Machines of support vectors in practice
- Keras.js – machine learning in the browser, you can touch the work of the algorithms of machine learning, helps in learning
- Advantages and disadvantages of AUC and accuracy
- Neural networks for transferring style to photos
- Style transfer using TensorFlow
- Ritchie Ng – a collection of resources for machine learning
- Overview of optimization methods by gradient descent in practice
- Lectures on support vector machines from the University of Utah
- Loss functions for the classification problem
