Artificial Intelligence: Can Computers Think? |
From inside the book
Results 1-3 of 11
Page 83
... unknown lever . If we have a fixed proba- bility , then with a very high probability we will be able to know whether the known probability is greater than the unknown one or less . If it is less we keep using the known lever for the ...
... unknown lever . If we have a fixed proba- bility , then with a very high probability we will be able to know whether the known probability is greater than the unknown one or less . If it is less we keep using the known lever for the ...
Page 84
... unknown proba- bility ? What are we going to average over ? What we assume is that the unknown probability , which we will call q , has a distribution function , a density function ( q ) dq , 0 ≤q≤ 1 . The simplest assumption will be ...
... unknown proba- bility ? What are we going to average over ? What we assume is that the unknown probability , which we will call q , has a distribution function , a density function ( q ) dq , 0 ≤q≤ 1 . The simplest assumption will be ...
Page 88
... UNKNOWN PARAMETER For example , we can say that not only do we have an unknown proba- bility distribution , but that the unknown probability has an unknown parameter . Let us give a very simple example of how this could occur . Suppose ...
... UNKNOWN PARAMETER For example , we can say that not only do we have an unknown proba- bility distribution , but that the unknown probability has an unknown parameter . Let us give a very simple example of how this could occur . Suppose ...
Common terms and phrases
actual algorithm analog computer answer applications approach approximate policies arithmetic artificial intelligence assumptions average outcome behavior Bellman BIBLIOGRAPHY AND COMMENTS bility chapter chess COMMENTS Section complex concept consider Control Processes criterion function dealer determine deterministic device difficulty digital computer discussed draw a card drug Dynamic Programming effect example expected gain experience experimentation fast storage feasible fifteen puzzle foregoing functional equation fuzzy sets human idea important instinct interesting large number learning levels logic machine mathe mathematical analysis mathematical problems mathematical theory mathematician matical mean method minimize minimum Monte Carlo Method multistage decision process observe operations optimal play optimal policy particular path patient pattern recognition player possible precise proba probability distribution probability theory procedure puter puzzles question reasonable simple situations solve space stochastic approximation stochastic process structure talk tion transformation uncertainty unknown probability York Zadeh