-1
$\begingroup$

This is my first Question so apologies if I do not stick to the standards.

What I want to understand is how is all of the following topics:

  1. Probability
  2. Different Probability Distributions.
  3. Baye's Theorem and soo on..

related to Machine Learning and Deep Learning modelling.

Like can someone give an example

  • How these theories help us developing let's say a simple Classification problem to a complex Neural Network that detects Faces for example?

  • Can someone pinpoint the exact stage at which we need to utilise these concepts and how do we actually use them?

$\endgroup$
3
  • 3
    $\begingroup$ Welcome to Data Science! This question is too broad to really fit the format of this Stack (or the statistics Stack). However, as you get into machine learning, one of the first topics you will encounter is the expected value, which is (typically) what a machine learning model aims to predict. Expected value is a property of a probability distribution. Perhaps you can read some introductory material on machine learning, preferably starting with basics like simple linear regression, and post more specific questions as they arise. $\endgroup$
    – Dave
    Commented Oct 19, 2022 at 14:37
  • $\begingroup$ @Dave what you mean when you say that a machine learning model aims to predict the expected values? I thought they are predicting the probabilities. $\endgroup$ Commented Oct 19, 2022 at 15:19
  • 1
    $\begingroup$ Depending on the situation, it might be that the expected value is a probability. Perhaps that could be a specific question to post: “Machine learning predicts the expected value? I thought it predicted probability!” $\endgroup$
    – Dave
    Commented Oct 19, 2022 at 15:24

1 Answer 1

1
$\begingroup$

Usage of Bayes Theorem example:

Let's say you build a object recognition model to recognize if person on a photo is doing a robbery. If he wears a mask, then the probability of him being a robber increases lets say to 50%.

Then, it makes sense for to add feature to your object detection model to recognize if such person is wearing a mask.

Then, based on multiple objects found on the photo (mask, crowbar, pistol, etc) the probability will grow further and further and you can think of building your model upon such set, which all together will give you good percentage of being a robber..

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.