Correction: (max 500 charact.)
The complaint will not be published.
I 252
Puzzle/Poundstone: anticipate the basic problem of inference, namely the question of how to recognize a paradox - (NP-complete).
>
Conclusions , >
Paradoxes , >
Recognition .
Right turn rule: is overcome by islands, therefore inefficient.
Solution: Tremaux: thread, at a dead end return to the last node. Also mark dead ends. - Two breadcrumbs mark old dead ends. - At old node choose a path that was not chosen before.
I 259
Results in first exploring remote areas.
I 267
"Problem of the longest path": is there an easy way? Trying does not lead directly to the shortest one. - No intelligent algorithm is available.
I 270
NP-Complete/Poundstone: the answers are easy to verify!
E.g. puzzle: the right way may only be two nodes away, but you had to try out many combinations.
>
Review , >
Verification , >
Confirmation .
I 282
Prove that NP problems cannot be solved with a computer.
I 274
Combination/Permutation/Combinatorics: P: polynomial function: n² - E.g. puzzle with 5000 parts. solvable - NP: exponential function. 2
n . E.g. Puzzle with 5000 paths - unsolvable.
In general: difficult to solve.
NP: "non-deterministically polynomial-temporally complete".
I 276
So far no evidence that NP problems cannot be solved in polynomial time. - But no empirical evidence. - The process of logical inferences is itself an NP problem. - Our conclusions about the world are limited.
I 281
The chain end, the very basis of our knowledge, can be recognized in polynomial time and checked for contradictions (lists; - but not walkable as a puzzle).
>
Knowledge , >
Contradictions , >
Consistency .