Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
32 Cards in this Set
- Front
- Back
A number between 0 and 1 that represents the relevance of an object to current processing. |
Activation |
|
a lack of decisiveness or commitment resulting from a failure to make a choice between alternatives. |
Ambiguity |
|
Learning algorithm in feedforward networks that adjusts the strengths of the links between neurons. |
Backpropogation |
|
emphasizes the importance of connections among simple neuronlike structures, but is also sometimes discussed in terms of neural networks or parallel distributed processing |
Connection |
|
where electrical signals flow into during connection through a synapse. |
Dendrite |
|
refers to the removal of ambiguity by making something clear |
Disambiguation |
|
a pattern of activity across many units that represents a concept. More neurologically plausible. |
Distributed Representation |
|
Neurons have ________ of neurotransmitters that provide chemical links between them, so the brain must be considered in chemical as well as electrical terms. |
Dozens |
|
Philosophical view that the mind consists of two separate substances, soul and body.
|
Dualism |
|
activity of one neuron that can increase the firing rate of another. |
Excitatory synapse |
|
Artificial neural network in which the flow of activity is in one direction, from input neurons to output neurons |
Feedforward |
|
Version of materialism according to which mental states are defined by their functional relations, not by any particular kind of physical realization. |
Functionalism |
|
Process in neural networks that strengthens the association between two neurons that are simultaneously active. |
Hebbian Learning |
|
units in the middle of the feedforward system as a bridge from input units to the output units. Do not receive or send units directly to or from the outside. |
Hidden Units |
|
activity of one neuron that decreases the firing rate of another. |
Inhibitory Synapse |
|
recieves information into a network, activates the network. |
Input units |
|
a representation in which neuron-like structures are given an identifiable interpretation in terms of specifiable concepts or propositions. Single unit. |
Local Representation |
|
Philosophical view that minds are purely physical. |
Materialism |
|
Use of language to understand and experience one kind of thing in terms of another |
Metaphor |
|
visual analogy for distributed processing. |
Octopus Network |
|
number of neurons in the brain. |
One hundred billion |
|
each neuron has about _______ synapses. |
one thousand |
|
send information outside the network |
Output units |
|
processing that simultaneously resolves parallel constraints. Process in which a problem is solved by using a parallel algorithm to find the best assignment of values to interconnected aspects of the problem. |
Parallel Constraint Satisfaction |
|
property of a system in which many activities take place at once |
Parallel processing |
|
claims that every mental state such as being conscious of the smell of donuts is a physical state of the brain. |
Reductive Materialism |
|
Process in which an artificial neural network reaches a state of stable activations. |
Relaxation |
|
neurons fire up a maximum of up to this number per second. |
Several hundred |
|
activation flowing back and forth between two units. |
Symmetric |
|
Space in which a signal passes from one neuron to another. |
Synapse |
|
an ordered set of numbers that indicates the activation of different units . |
Vector |
|
a number that indicates how important one unit is to the activity of another object. |
Weight |