"A neural network is a massively parallel distributed processor that has a natural propensity for storing experiential knowledge and making it available for use. It resembles the brain in two respects: 1. Knowledge is acquired by the network through a learning process. 2. Interneuron connection strengths known as synaptic weights are used to store the knowledge." (Igor Aleksander, "An introduction to neural computing", 1990)
"Neural Computing is the study of networks of adaptable nodes which through a process of learning from task examples, store experiential knowledge and make it available for use." (Igor Aleksander, "An introduction to neural computing", 1990)
"A fuzzy cognitive map or FCM draws a causal picture. It ties facts and things and processes to values and policies and objectives. And it lets you predict how complex events interact and play out. [...] Neural nets give a shortcut to tuning an FCM. The trick is to let the fuzzy causal edges change as if they were synapses in a neural net. They cannot change with the same math laws because FCM edges stand for causal effect not signal flow. We bombard the FCM nodes with real data. The data state which nodes are on or off and to which degree at each moment in time. Then the edges grow among the nodes." (Bart Kosko, "Fuzzy Thinking: The new science of fuzzy logic", 1993)
"A neural net acts like the eyes and ears of an adaptive fuzzy system, a fuzzy system whose rules change with experience. The neural net senses fuzzy patterns in the data and learns to associate the patterns. The associations are rules: If fuzzy set A, then fuzzy set B." (Bart Kosko, "Fuzzy Thinking: The new science of fuzzy logic", 1993)
"A neural network is characterized by A) its pattern of connections between the neurons (called its architecture), B) its method of determining the weights on the connections (called its training, or learning, algorithm), and C) its activation function." (Laurene Fausett, "Fundamentals of Neural Networks", 1994)
"An artificial neural network is an information-processing system that has certain performance characteristics in common with biological neural networks. Artificial neural networks have been developed as generalizations of mathematical models of human cognition or neural biology, based on the assumptions that: 1. Information processing occurs at many simple elements called neurons. 2. Signals are passed between neurons over connection links. 3. Each connection link has an associated weight, which, in a typical neural net, multiplies the signal transmitted. 4. Each neuron applies an activation function (usually nonlinear) to its net input (sum of weighted input signals) to determine its output signal." (Laurene Fausett, "Fundamentals of Neural Networks", 1994)
"At the other far extreme, we find many systems ordered as a patchwork of parallel operations, very much as in the neural network of a brain or in a colony of ants. Action in these systems proceeds in a messy cascade of interdependent events. Instead of the discrete ticks of cause and effect that run a clock, a thousand clock springs try to simultaneously run a parallel system. Since there is no chain of command, the particular action of any single spring diffuses into the whole, making it easier for the sum of the whole to overwhelm the parts of the whole. What emerges from the collective is not a series of critical individual actions but a multitude of simultaneous actions whose collective pattern is far more important. This is the swarm model." (Kevin Kelly, "Out of Control: The New Biology of Machines, Social Systems and the Economic World", 1995)
"Many of the basic functions performed by neural networks are mirrored by human abilities. These include making distinctions between items (classification), dividing similar things into groups (clustering), associating two or more things (associative memory), learning to predict outcomes based on examples (modeling), being able to predict into the future (time-series forecasting), and finally juggling multiple goals and coming up with a good- enough solution (constraint satisfaction)."
"More than just a new computing architecture, neural networks offer a completely different paradigm for solving problems with computers. […] The process of learning in neural networks is to use feedback to adjust internal connections, which in turn affect the output or answer produced. The neural processing element combines all of the inputs to it and produces an output, which is essentially a measure of the match between the input pattern and its connection weights. When hundreds of these neural processors are combined, we have the ability to solve difficult problems such as credit scoring." (Joseph P Bigus,"Data Mining with Neural Networks: Solving business problems from application development to decision support", 1996)
"When training a neural network, it is important to understand when to stop. […] If the same training patterns or examples are given to the neural network over and over, and the weights are adjusted to match the desired outputs, we are essentially telling the network to memorize the patterns, rather than to extract the essence of the relationships. What happens is that the neural network performs extremely well on the training data. However, when it is presented with patterns it hasn't seen before, it cannot generalize and does not perform well. What is the problem? It is called overtraining." (Joseph P Bigus,"Data Mining with Neural Networks: Solving business problems from application development to decision support", 1996)
"A neural network consists of large numbers of simple neurons that are richly interconnected. The weights associated with the connections between neurons determine the characteristics of the network. During a training period, the network adjusts the values of the interconnecting weights. The value of any specific weight has no significance; it is the patterns of weight values in the whole system that bear information. Since these patterns are complex, and are generated by the network itself (by means of a general learning strategy applicable to the whole network), there is no abstract procedure available to describe the process used by the network to solve the problem. There are only complex patterns of relationships."
"The main "Neural nets have no central control in the classical sense. Processing is distributed over the network and the roles of the various components (or groups of components) change dynamically. This does not preclude any part of the network from developing a regulating function, but that will be determined by the evolutionary needs of the system." (Paul Cilliers, "Complexity and Postmodernism: Understanding Complex Systems", 1998)
"Neural networks are computer structures, based on the way in which the brain functions. They are used to recognize complex patterns. They typically involve a number of interconnected elements that are used to create a type of "model" of some pattern or phenomenon. The model is formed as a function of the 'weights,' or strengths, of the connections between the elements in the network. This inner 'model' determines the output of the network." (Robert B Dilts, "Modeling with NLP", 1998)
"Neural networks conserve the complexity of the systems they model because they have complex structures themselves. Neural networks encode information about their environment in a distributed form. […] Neural networks have the capacity to self-organise their internal structure." (Paul Cilliers, "Complexity and Postmodernism: Understanding Complex Systems", 1998)
"The internal structure of a connectionist network develops through a process of self-organisation, whereas rule-based systems have to search through pre-programmed options that define the structure largely in an a priori fashion. In this sense, learning is an implicit characteristic of neural networks. In rule-based systems, learning can only take place through explicitly formulated procedures." (Paul Cilliers, "Complexity and Postmodernism: Understanding Complex Systems", 1998) that hampered the development of neural network models was the absence of a mathematical model for adjusting the weights of neurons situated somewhere in the middle of the network, and not directly connected to the input or output." (Paul Cilliers, "Complexity and Postmodernism: Understanding Complex Systems", 1998)
"Whereas formal systems apply inference rules to logical variables, neural networks apply evolutive principles to numerical variables. Instead of calculating a solution, the network settles into a condition that satisfies the constraints imposed on it." (Paul Cilliers, "Complexity and Postmodernism: Understanding Complex Systems", 1998)
No comments:
Post a Comment