Typical algorithm of AI (artificial intelligence)
We will explain typical algorithms such as "neural networks," "genetic algorithms," and "expert systems," which are also called AI three families, which play an important role in AI research.
neural network
Neural networks are AI models of the structure and function of neurons (nerve cells that make up the brain of living organisms). When a neuron receives an electric signal above a certain value from another neuron, it gets excited and sends an electric signal to the neuron connected to that point. This is a numerical model of the mechanism of cooperative behavior between neurons.
A neural network consists of an input layer that stores data, an intermediate layer (hidden layer) that processes weights flowing from the input layer, and an output layer that outputs results. When a human becomes a teacher and teaches a set of examples and model answers (teacher signals) to a neural network, the neural network itself will be able to judge and infer even the range that is not taught.
Genetic algorithm
The genetic algorithm is an AI with Darwin's theory of evolution as a motif.
The following is a summary of Darwin's theory of evolution.
As this "excellent individual" = "good answer", the genetic algorithm tries to derive the optimum solution using evolutionary methods.
The best thing about genetic algorithms is to find the best answer from the huge number of combinations. You can quickly find the optimal solution for a problem that causes a combinatorial explosion at a level that is difficult to calculate manually.
Expert system
The expert system is an AI modeled on the human "idea". Unlike other AI models, there is no mechanism for learning on your own.
First, we will hear about the situations that can be considered by a specific expert and how to deal with them, judgments, and predictions, and then define the rules based on them. Based on the rules set there, we will determine which situation the user's inquiry applies to, and make the defined judgments and predictions.
As shown in the examples below, it is especially useful in diagnosing diseases in the medical field.
As mentioned above, we will make a diagnosis prepared in advance according to the response from the user.
the more accurate it will be, but if you have too many rules, it can be difficult to make each rule consistent. Furthermore, if there are omissions or omissions in important rules, it will not be possible to make a correct judgment.
It is also a concern of the expert system that setting rules requires the help of an expert, and even if the rules can be set correctly, it is not possible to derive more answers than experts.
