Conditional Probability

Updating beliefs with new evidence

🎯 Definition

The probability of event \(E\) occurring, given that event \(F\) has already occurred.

\( P(E|F) = \frac{P(E \cap F)}{P(F)} \)

"The part of E that is inside F, divided by the total size of F"

🔄 Bayes' Theorem

Bayes' Theorem allows us to flip conditional probabilities. It's crucial for medical diagnosis, spam filtering, and AI.

\( P(A|B) = \frac{P(B|A)P(A)}{P(B)} \)

Interactive: Medical Test Paradox

A test is 99% accurate. You test positive. What is the chance you actually have the disease? (Hint: It depends on how rare the disease is!)

1%
99%
5%

Probability you have disease given Positive Test:

16.6%

Even with high accuracy, if the disease is rare, false positives can outnumber true positives!

🚪 The Monty Hall Problem

Behind one door is a Car 🚗. Behind the others, Goats 🐐. You pick a door. Monty opens another door with a Goat. Should you switch?

1
2
3

Pick a door!

Stay Wins

0

Switch Wins

0

📝 Test Your Understanding

Question 1:

If \(A\) and \(B\) are independent, what is \(P(A|B)\)?

Previous: Information Theory Back to Syllabus