Artificial IntelligenceMcGraw-Hill, 1991 - 621 páginas |
Dentro del libro
Resultados 1-3 de 15
Página 510
... active is given by p : 1 1 + AE / T where AE is the sum of the unit's active input lines and T is the " temperature " of the network . Stochastic updating of units is very similar to updating in Hopfield nets , except for the ...
... active is given by p : 1 1 + AE / T where AE is the sum of the unit's active input lines and T is the " temperature " of the network . Stochastic updating of units is very similar to updating in Hopfield nets , except for the ...
Página 511
... active for any given input . One solution to this problem is to let the network settle , find the output unit with the highest level of activation , and set that unit to 1 and all other output units to 0. In other words , the output ...
... active for any given input . One solution to this problem is to let the network settle , find the output unit with the highest level of activation , and set that unit to 1 and all other output units to 0. In other words , the output ...
Página 513
... active according to some natural division of the inputs . 1. Present an input vector , denoted ( X1 , X2 , ... , Xn ) ... active . " 4. Adjust the weights on the input lines that lead to the single active output unit : Xj Aw ; = n nw for ...
... active according to some natural division of the inputs . 1. Present an input vector , denoted ( X1 , X2 , ... , Xn ) ... active . " 4. Adjust the weights on the input lines that lead to the single active output unit : Xj Aw ; = n nw for ...
Contenido
Weak SlotandFiller Structures | 9 |
6 | 24 |
Heuristic Search Techniques | 63 |
Derechos de autor | |
Otras 24 secciones no mostradas
Otras ediciones - Ver todas
Términos y frases comunes
Abbott algorithm answer apply approach Artificial Intelligence assertions attributes axioms backpropagation backtracking backward backward reasoning belief best-first search breadth-first search Cabot Caesar Chapter clauses concept consider constraints contains contexts contradiction corresponding define depth-first depth-first search described discussed domain example explicitly fact given goal graph heuristic heuristic function Horn clauses important inference inheritance input instance interpretation justification knowledge base knowledge representation labeled learning logical assertions Marcus match move MYCIN node nonmonotonic reasoning object operators particular path perceptron possible preconditions predicate logic problem problem-solving procedure produce production system PROLOG propagation propositional logic question represent resolution result robot rules Section semantic semantic net sentence shown in Figure simple slot solution solve space specific statements step strategy structure Suppose suspect syntactic task techniques theorem things tree true truth maintenance system variables wff's