Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- 1.There is an algorithm that when called returns either 0 or 1 ( for heads or tails ).
- 2. The algorithm can call a special function CoinFlip() which picks a truly random number ( 0 or 1 ).
- 3. The algorithm can remember state between each invocation.
- 4. The current score for the algorithm is the number of times it has been called, minus the number of times it has called CoinFlip()
- Q1) Given a sequence of N results returned by the algorithm, and providing that we pick N to be large enough, can we always determine if its current score is likely to be greater than 0 ?
- Q2) Is there an algorithm, which will increase itβs score as time goes on, but can not be distinguished (from looking at its results ) from an algorithm that will always have a score of 0.
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement