Understanding Bernoulli Trials, Bayesian Setting

Question

I am required to complete a project on ML applications. I guess there is a lot of statistics in ML, not helpful for a non-maths background.

I am getting too bogged down by notations. There are too many notations.

I am trying to read about what Bernoulli trials are and I can't relate to it.

What is a Bayesian setting and why is Bayesian thing everywhere? What makes it such an omnipotent distribution?

Are there any theorems/theories that I must know ? Any book/notes/resources where I could learn about these stuff in a relatable way (I have very little maths background, but I can learn)?

Answer 1

What is a Bayesian setting and why is Bayesian thing everywhere?

In very simple terms:

Bayesian is a statistical setting, where the likelihood of an event happening (called the posterior) depends on the prior trials or observations (called the prior(s)).

Bayesian networks is an extension of the above, forming a chain or a network of inferencing.

Are there any theorems/theories that I must know?

For understanding the Bayesian paradigm, you need to know the Bayesian theorem/relation, which is basically:

$$P(\theta|d) = \dfrac{P(d|\theta)P(\theta)}{P(d)}.$$

Any book/notes/resources where I could learn about these stuff in a relatable way (I have very little maths background, but I can learn)?

I would highly recommend "Doing Bayesian Analysis" by John Krushke