Hassan Ijaz

Ai, Web & Design
← Back to all topics
Probability FundamentalsTopic 6 of 58

Bayes' theorem and Bayesian inference

Medical test simulator where users adjust disease prevalence and test accuracy sliders to see how posterior probabilities change, with visual prior/posterior distributions

Concept Overview

Bayes' theorem is a fundamental result in probability theory that describes how to update beliefs in light of new evidence. It forms the foundation of Bayesian inference and has applications from medical diagnosis to machine learning.

The Formula

P(A|B) = (P(B|A) × P(A))/ P(B)

P(A|B): Posterior probability - What we want to find

P(B|A): Likelihood - How likely is the evidence given our hypothesis

P(A): Prior probability - Our initial belief before seeing evidence

P(B): Marginal likelihood - Total probability of the evidence

Alternative Forms

P(B) = P(B|A) × P(A) + P(B|A') × P(A')

Law of Total Probability - useful when P(B) is hard to calculate directly

P(A|B) ∝ P(B|A) × P(A)

Proportional form - often used when comparing hypotheses

Medical Testing Example

Consider a disease test with:

  • Disease prevalence: 1% (prior)
  • Test sensitivity: 99% (true positive rate)
  • Test specificity: 95% (true negative rate)

If someone tests positive, what's the probability they have the disease?

P(Disease|Positive) = 0.99 × 0.01 / (0.99 × 0.01 + 0.05 × 0.99) ≈ 16.7%

Surprisingly low! This is because the disease is rare, so most positives are false positives.

Key Insights

  • Base Rate Matters:Rare conditions remain unlikely even with positive tests
  • Evidence Strength Varies:Some evidence is more informative than others
  • Sequential Updates:Can apply Bayes' theorem multiple times as new evidence arrives
  • Prior Selection Matters:Different priors can lead to different conclusions

Common Applications

Medical Diagnosis

Updating disease probability based on test results and symptoms

Spam Filtering

Classifying emails based on word frequencies

Machine Learning

Naive Bayes classifiers, Bayesian neural networks

A/B Testing

Bayesian approaches to experiment analysis

Remember: Bayes' theorem is about updating beliefs rationally. The posterior combines what we knew before (prior) with what the data tells us (likelihood).

Use the medical test simulator below to see how changing disease prevalence and test accuracy affects the posterior probability. Watch how the visual prior and posterior distributions change as you adjust the parameters.

Interactive Visualization

Loading interactive visualization...

Medical test simulator where users adjust disease prevalence and test accuracy sliders to see how posterior probabilities change, with visual prior/posterior distributions