Why do we learn probability theories for machine learning? - Part One.
A large number of undergraduate students as well as graduate students are interested in machine learning or deep learning. It sounds allure to become an AI Engineer. But the first thing first, most of the students who are from a non-computer science background have a lack of proper allusion to learn AI. By the way, our post is about machine learning. But why am I talking about AI?
Well, Machine learning is a sub-domain of Artificial Intelligence. Where we aim to train our systems to predict intelligence outcomes by applying statistical and probabilistic theories and algorithms. Learning “machine learning” is not alacrity when it comes to Mathematics. But then again, the fundamental rule to learn “machine learning” is to have extrinsic knowledge of probability, statistics and gradient calculus.
Some of you ended up with some online courses to learn machine learning. But in a long fidelity, it requires a sound knowledge of theories. Rather than having practical deployment skills. But sadly, most of the courses do not compile foundational theories.
Most of the top-tier companies like Google, Microsoft, Amazon, Facebook require their Data Scientists, Machine Learning Engineers, Deep Learning Engineers to have a PhD from a top university across the world. Can’t believe? Let’s check out the link. I guess you already get some better intuitions from it. Remember, preferred qualifications always count. But to get a PhD in Machine Learning, all you need to have proficient knowledge in theories.
A lot of expostulations till now, so let’s jumped into our main business.
What is probability?
Probability is the branch of mathematics concerning numerical descriptions of how likely an event is to occur or how likely it is that a proposition is true. So what’s the coherent part of probability in machine learning then?
To get the answer, wait until part two comes.