The development history and characteristics of artificial neural networks

The artificial neural network ANN is a research hotspot in the field of artificial intelligence since the 1980s. The artificial neural network ANN is simply referred to as a neural network or a neural network. Deep learning is actually a deep neural network DNN, that is, deep learning is developed from the artificial neural network ANN model, so it is necessary to further explore the artificial neural network ANN. ^_^

In the past ten years, the research work of artificial neural network ANN has been deepened and has made great progress. It has successfully solved many modern technologies in the fields of pattern recognition, intelligent robot, automatic control, predictive estimation, biology, medicine, economy and so on. The practical problems that are difficult to solve by computers show good intelligence.

So what exactly is an artificial neural network ANN?

The artificial neural network ANN abstracts the human brain neural network from the perspective of information processing, establishes a simple model, and forms different networks according to different connection methods. The artificial neural network ANN is an operational model consisting of a large number of nodes (or neurons) connected to each other. Each node (neuron) represents a specific output function called the excitation function (acTIvaTIon funcTIon). The connection between every two nodes represents a weighting value for passing the connection signal, called weight, which is equivalent to the memory of the artificial neural network. The output of the network varies depending on the connection method of the network, the weight value and the excitation function. The network itself is usually an approximation of an algorithm or function in nature, or it may be an expression of a logic strategy.

The development of artificial neural network ANN:

1) The concept of artificial neural network ANN is by W. S. McCulloch and W. Pitts et al. were proposed in 1943. They proposed formal mathematical descriptions and network structure methods of neurons through the MP model.

2) In the 1960s, artificial neural networks were further developed, and a more complete neural network model was proposed, including perceptrons and adaptive linear components. 1982, J. J. Hopfield proposed the Hopfield neural mesh model and introduced the concept of "computational energy" to give a judgment of network stability. In 1984, he proposed a continuous-time Hopfield neural network model, which opened up a new way for neural networks to use associative memory and optimize computation. This pioneering research work has strongly promoted the research of neural networks.

3) In 1985, some scholars proposed the Boltzmann model, and used statistical thermodynamic simulated annealing technology to ensure that the whole system tends to be globally stable.

4) In 1986, the study of cognitive microstructure was carried out, and the theory of parallel distributed processing was proposed.

5) In 1986, Rumelhart, Hinton, Williams developed the BP algorithm. To date, BP algorithms have been used to solve a large number of practical problems.

6) In 1988, Linsker proposed a new self-organization theory for perceptron networks, and formed the largest mutual information theory based on Shanon information theory, thus igniting the ray of information application theory based on NN.

7) In 1988, Broomhead and Lowe proposed a hierarchical network design method using Radial basis funcTIon (RBF) to link the design of NN with numerical analysis and linear adaptive filtering.

8) In the early 1990s, Vapnik et al. proposed the concepts of support vector machines (SVM) and VC (Vapnik-Chervonenkis) dimensions.

9) The US Congress passed a resolution to define the decade beginning on January 5, 1990 as a “brain decade.” The International Research Organization called on its member states to turn “the decade of the brain” into a global act. In the "Real World Computing (RWC)" project in Japan, the study of artificial intelligence has become an important component. The research of artificial neural networks has been paid attention to by various developed countries.

Artificial neural network features:

Artificial neural network is a nonlinear, adaptive information processing system composed of a large number of processing unit interconnections. It is based on the results of modern neuroscience research, trying to process information by simulating the way the brain neural network processes and memorizes information. Artificial neural networks have four basic characteristics:

(1) Nonlinearity: Artificial neurons are activated or suppressed in two different states, and this behavior is mathematically represented as a nonlinear relationship. A network of thresholded neurons has better performance and can improve fault tolerance and storage capacity.

(2) Non-limiting: A neural network is usually made up of a wide range of neurons. The overall behavior of a system depends not only on the characteristics of individual neurons, but also on the interactions and interconnections between the elements. Simulate the non-limiting nature of the brain through a large number of connections between units.

(3) Very qualitative: artificial neural networks have adaptive, self-organizing, self-learning capabilities. The neural network can not only change the information processed, but also the nonlinear dynamic system itself is constantly changing while processing the information. An iterative process is used to describe the evolution of the dynamic system.

(4) Non-convexity: The evolution direction of a system will depend on a particular state function under certain conditions. Non-convexity means that this function has multiple extreme values, so the system has multiple stable equilibrium states, which will lead to the diversity of system evolution.

Solar Lights

Solar lighting systems use solar cells to convert solar energy in nature into electricity and provide it to LED light sources. Due to the low voltage, energy saving and long-term characteristics of LED light sources, the application of solar LED lighting systems will achieve high energy efficiency, reliability, and practical value.

Solar panels are used to receive solar radiation during the day and convert light energy intoelectrical energy stored in the storage Battery by controller while the battery starts discharging and illuminate the light source in the evening. In the morning, the controller will switch off the battery and stop discharging according to the illuminance. The main function of the controller is to control the street lamp switch, adjust the brightness, and protect the battery to prolong the service life.

Solar Lights,Solar Lights,Led Lights,Garden Decoration, Led Street Lights,Led Lamp

Wuxi Sunket New Energy Technology Co.,Ltd , https://www.sunketsolar.com

Posted on