Neural Networks
With most computer games over the year people have only been given the illusion of intelligence, as we are not yet developed enough to create actual intelligence in our games. It is extremely difficult to plagiarize human prediction when it comes to games. In some cases humans can be extremely predictable e.g. In the game of American Football some teams always throw the ball where as other teams always run with the ball. In some cases however humans can be extremely unpredictable e.g.The way of unpredictable thinking is extremely difficult to simulate in a computer, as a computer cannot think for itself. When it comes to playing games with a computer the computer itself is not actually thinking it is programmed with all the predicted moves that a user can make. The programmer of the game actually sits down and programs all the moves that take place if a user does something. This then gives the user the illusion that the computer is coming up with way to try and beat them. This is where a new way of thinking came in. The idea was to actually simulate the way a human thinks by actually creating a computer simulated brain. This in theory anyway would allow the program to actually think and learn through trial and error. . . .
· The activity of each hidden unit is determined by the activities of the input units and the weights on the connections between the input and the hidden units. One of the most influential researchers into neurological systems was Donald Hebb who proposed that learning consisted principally in altering the "strength" of synaptic connections. The weights between the input and hidden units determine when each hidden unit is active, and so by modifying these weights, a hidden unit can choose what it represents. A unit uses an input-output function that transforms the total input into the outgoing activity. We present the network with training examples, which consist of a pattern of activities for the input units together with the desired pattern of activities for the output units. A neuron fires only if the total signal received at the cell body from the dendrites exceeds a certain level (the firing threshold). It became rapidly apparent that these systems, although very useful in some domains, failed to capture certain key aspects of human intelligence. Threshold units: The output is set at one of two levels, depending on whether the total input is greater than or less than some threshold value. The main branch of Artificial Intelligence research in the 1960s -1980s produced Expert Systems. Sigmoid units: The output varies continuously but not linearly as the input changes. Sigmoid units bear a greater resemblance to real neurones than do linear or threshold units, but all three must be considered rough approximations. The connections determine whether it is possible for one unit to influence another. The strength of the signal received by a neuron (and therefore its chances of firing) critically depends on the efficacy of the synapses. Neural networks are typically composed of interconnected "units", which serve as model neurones.
Common topics in this essay:
Donald Hebb, Sigmoid Linear, Expert Systems, American Football, , Artificial Intelligence, hidden units, total input, input units, neural networks, output units, neural network, units connected, hidden unit, units output, input hidden units, artificial intelligence, units connected layer, activities input units, |
Acceptance Essays
