Things that I find interesting, mostly about nature, science, mathematics, programming, and Thailand's fight against corruption.
Showing posts with label Bayesian Inference. Show all posts
Showing posts with label Bayesian Inference. Show all posts
Wednesday, April 02, 2014
หวยล็อก รถนายก ความน่าจะเป็น และ Bayesian Inference
มีเพื่อนใน Facebook แชร์ลิงค์จาก ThaiPublica ว่าเลขทะเบียนรถนายกตรงกับหวย 8 งวดจาก 63 งวด (ทั้งหมดเป็นเลขท้ายสองและสามตัว หรือเลขท้ายรางวัลที่ 1) แปลว่าน่าจะมีการล็อคหรือไม่
ผมก็ไม่แน่ใจว่ามีการล็อคหรือไม่ แต่คิดว่าอาจจะเป็นความบังเอิญเพราะมีรถที่เกี่ยวข้องหลายคันก็ได้ ยกตัวอย่างเช่น ถ้ามีรถเป็นร้อยคัน ก็น่าจะมีเลขทะเบียนสักอันตรงกับหวยเกือบทุกงวด จึงทดลองคำนวณดูว่าความน่าจะเป็นต่างๆที่เกี่ยวข้องว่ามากน้อยอย่างไร และพบว่าถ้าไม่มีการล็อคเลขหวย เราสามารถใช้ Bayesian Inference ประมาณจำนวนรถที่นายกใช้ว่ามีประมาณ 6 ± 2 คัน
รายละเอียดการคำนวณอยู่ที่ http://www.atriumtech.com/ko/mathematica/PooLottery/ ครับ
Saturday, January 19, 2008
Applied Probability: How To Write A Spelling Corrector In 21 Lines Of Python
Once again Bayes' Theorem comes in handy. This time, Peter Norvig shows us how to make a not-so-bad spelling corrector. It was done in 21 lines of easy-to-understand Python! I actually could understand it in one pass of reading. (It usually takes me many reading passes to understand any algorithm--I'm dense that way.)
If you have children, they should have some basic math and probability skills. Rational thoughts need those skills. Otherwise, they won't be able to make quantitative comparisons.
If you have children, they should have some basic math and probability skills. Rational thoughts need those skills. Otherwise, they won't be able to make quantitative comparisons.
Sunday, August 12, 2007
Sometimes Experts Don't Know Much
The Rosenhan experiment showed that psychiatric hospital staff couldn't reliably distinguish between the sane and the insane in the hospital. The full paper can be found here.
According to another study at Cornell, you can implant false memories into preschool children to make them think the memories are true, and experts cannot distinguish, by interviewing the children, which memories are true and which are false.
So, when experts' claims conflict with reality, you should be careful in checking whether the claims are true or not. Most people (yes, including me) appear (or pretend) to know more than they actually know.
--- - ---- - ----- ---------
P.S. How to believe anything? That's a hard question that cannot be completely solved. I think we can only hope to converge closer and closer to the truth. Gotama Buddha tried to teach the Kalamas people using the list we know as Kalama Sutra. Personally, I like the Bayesian approach of updating my belief on whether something is true or false as I gain more and more evidences. A short overview of Bayesian inference is here. A reading list can be found here.
According to another study at Cornell, you can implant false memories into preschool children to make them think the memories are true, and experts cannot distinguish, by interviewing the children, which memories are true and which are false.
So, when experts' claims conflict with reality, you should be careful in checking whether the claims are true or not. Most people (yes, including me) appear (or pretend) to know more than they actually know.
--- - ---- - ----- ---------
P.S. How to believe anything? That's a hard question that cannot be completely solved. I think we can only hope to converge closer and closer to the truth. Gotama Buddha tried to teach the Kalamas people using the list we know as Kalama Sutra. Personally, I like the Bayesian approach of updating my belief on whether something is true or false as I gain more and more evidences. A short overview of Bayesian inference is here. A reading list can be found here.
Subscribe to:
Posts (Atom)
