Skip to main content

Updating beliefs based on new evidence. How?


Bayes' Theorem is a way to update our beliefs about something based on new evidence. Here's how it works in simple terms:

Imagine you have a belief or a guess about how likely something is to happen. This is your prior probability.

Now, let's say you get some new information or evidence. Bayes' Theorem helps you figure out how this new evidence should change your original guess. Here’s the process:

  1. Prior Probability: This is what you think the chances are before you see any new evidence.
  2. Likelihood: This is how likely it is to see this new evidence if your initial guess were correct.
  3. Evidence: This is the probability of seeing this evidence under any circumstances, not just if your guess is correct.
  4. Posterior Probability: After considering the new evidence, this is your updated belief or guess.

The formula looks like this:

P(A|B) = \frac{P(B|A) \times P(A)}{P(B)}
  • P(A|B) is the probability of A given that B has occurred (your updated belief).
  • P(B|A) is the probability of B given that A is true (likelihood).
  • P(A) is the initial probability of A (prior probability).
  • P(B) is the total probability of B occurring (evidence).

Here’s a practical example:

  • Suppose you're trying to guess if it will rain tomorrow. Your prior probability might be 30% based on the weather forecast.
  • Then, you see dark clouds in the sky (new evidence).
  • You know from past experience that when it rains, there's an 80% chance of seeing dark clouds (P(B|A)), but dark clouds can appear even if it doesn't rain, say 20% of the time (P(B)).
  • Using Bayes' Theorem, you calculate how this new evidence changes your prediction about rain:

    • P(A|B) = (0.80 * 0.30) / 0.20 = 1.2 (but since probabilities can't exceed 1, this indicates you've significantly increased your belief in rain - you'd need to normalize this with all possible outcomes).

So, your updated chance of rain tomorrow, considering the dark clouds, would be much higher than 30%.

In essence, Bayes' Theorem is about refining your predictions with new data, making your guesses smarter and more accurate over time.

Comments

Post a Comment

Popular posts from this blog

Helen Mirren once said: Before you argue with someone, ask yourself.......

Helen Mirren once said: Before you argue with someone, ask yourself, is that person even mentally mature enough to grasp the concept of a different perspective. Because if not, there's absolutely no point. Not every argument is worth your energy. Sometimes, no matter how clearly you express yourself, the other person isn’t listening to understand—they’re listening to react. They’re stuck in their own perspective, unwilling to consider another viewpoint, and engaging with them only drains you. There’s a difference between a healthy discussion and a pointless debate. A conversation with someone who is open-minded, who values growth and understanding, can be enlightening—even if you don’t agree. But trying to reason with someone who refuses to see beyond their own beliefs? That’s like talking to a wall. No matter how much logic or truth you present, they will twist, deflect, or dismiss your words, not because you’re wrong, but because they’re unwilling to see another side. Maturity is...
Post a Comment
Read more

The battle against caste: Phule and Periyar's indomitable legacy

In the annals of India's social reform, two luminaries stand preeminent: Jotirao Phule and E.V. Ramasamy, colloquially known as Periyar. Their endeavours, ensconced in the 19th and 20th centuries, continue to sculpt the contemporary struggle against the entrenched caste system. Phule's educational renaissance Phule, born in 1827, was an intellectual vanguard who perceived education as the ultimate equaliser. He inaugurated the inaugural school for girls from lower castes in Pune, subverting the Brahminical hegemony that had long monopolized erudition. His Satyashodhak Samaj endeavoured to obliterate caste hierarchies through radical social reform. His magnum opus, "Gulamgiri" (Slavery), delineated poignant parallels between India's caste system and the subjugation of African-Americans, igniting a discourse on caste as an apparatus of servitude. Periyar's rationalist odyssey Periyar, born in 1879, assumed the mantle of social reform through the Dravidian moveme...
Post a Comment
Read more

AI & Higher Education: The Empty Classroom

  ARTIFICIAL INTELLIGENCE & HIGHER EDUCATION The Empty Classroom When students outsource learning to AI and companies cut the engineers who know better, both ends of the talent pipeline fray at once. India is not watching from a safe distance. Chuppala Nagesh Bhushan At the University of California, Berkeley, something unremarkable happened in spring 2026: a professor held office hours. The unremarkable part was that nobody came. Dan Garcia, who teaches CS 10, a broad introductory computing course popularly called “The Beauty and Joy of Computing,” found his calendar conspicuously clear at the very moment his gradebook became conspicuously alarming. Of the students who sat CS 10’s final examination, 35.3% received an F—five times the historical norm of roughly 7%. Two other courses in Berkeley’s elite Electrical Engineering and Computer Sciences department suffered similarly: 10.6% of CS 61A students failed, and 16.8% of those in EECS 127, an upper-division optimi...
1 comment
Read more