What is a good starting point in data science learning process? Traditionally we start with theories, then the mechanics and finally make hands dirty. This rapidly kills excitement by exponentially loosing interest, and at some point get out of track. This happens more often then not before doing any of the problem solving work we are really interested in. Is this the most efficient way to learn something?
I believe the best way for a learner to learn in data science is by examples, rather than by grammars, theories and rules. Little kids do not learn a new language by rules. If you want to learn how to swim, just go to a pool instead of searching a “how to” video on YouTube. The whole world of machine learning is about examples – train models with lots & lots of examples and then let the model solve a real problem. Humans have figured out this is the most efficient approach for machines to learn instead of giving them rules (because every rule has an exception, right?). This is how human’s learning process should also work. Once you dive into the technical analytical part with whatever tools/resources you have and you will automatically get interested in the mechanics of it, learn theories, while finding answers to curious questions.
To be clear, this is not to say theories are more or less important. This is just to say that – between theory and practice – a learner should start with practice, theory will just fall in line in the process.