# Solution 69: States, Expected Values, Ants, and an Octahedron

Share
Embed
• Published on Dec 6, 2018
• Let's follow the ants until at least two of them meet on the same vertex.
Congratulations to Aaron Lamoreaux, Max Haibara, adandap, Alex Edwards, and mstmar for successfully solving the last week's math challenge question! Aaron Lamoreaux was the first person to solve the question.
Your support is truly a huge encouragement.
Every subscriber and every like are wholeheartedly appreciated.
For more Weekly Math Challenges:

• Master mystery 4 months ago

Will this method even work if we need to find expected number of moves until all 3 ants land on the same vertex. Is it that only the calculated probabilities will change but the rest will remain the same??

• Chris X 5 months ago

Nice video, but I got a little confused with the mu1, mu2... I liked the problem though and I just solved it here usclip.net/video/mFZZPRal5fw/video.html using just programming (so not exact solution). I hope you liked it. I added this video's url as a reference to the problem :)

• DerivativeOfSenpi 6 months ago +1

Nice

• Michael Empeigne 6 months ago

what is the sum of the interior angles of a octohedron ?

• Datta Subrahmanyam 6 months ago +1

Can you please show step by step solution for 'Monty hall's problem ' for calculating probability of Win ?

• adandap 6 months ago +2

@Datta Subrahmanyam Well, no. The change of door is very much *not* about independence of events because the host cannot choose at random - his choice is restricted in 2/3 of cases. So the a posteriori probability is not 1/2, it's 2/3. You might also like to read about 'the principle of restricted choice' in contract bridge, which depends on the same mathematical reasoning. en.wikipedia.org/wiki/Principle_of_restricted_choice

• Datta Subrahmanyam 6 months ago

@adandap here it looks like a bernoulli's estimation of 3 independent events & with in each independent event, there are dependent events so the posteriori probability becomes 1/2 instead of 2/3, in 2nd iteration (after revealing one of the three options)

• adandap 6 months ago +2

@Datta Subrahmanyam I think that is the explanation. The a priori probability is 1/3 and the a posteriori probablity is 2/3. (Or 1/00 and 99/100 in the example at the end.) What more is there to it? Some problems just aren't that sophisticated...

• Datta Subrahmanyam 6 months ago

@adandap well thank you. I'm expecting a different explaination rather than just naive explaination.

Lots of places to find this done. Here's the numberphile version: usclip.net/video/4Lb-6rxZxx0/video.html

• matyourin 6 months ago +2

Complicated way to approach it... I solved it using a Markov process and calculating the Eigenvalues of the transformation matrix...seemed to be an easier approach to me.

• matyourin 6 months ago +2

As far as I know, historically, it was a similar problem that was given by a western university professor when everybody solved it in a more "counting orientied way" when Markov as an exchange student or something of that kind came up with his approach which was rather unfamiliar outside of Russian mathematic schools. So history sort of repeated itself here ;)