What is Smart
Before the introduction of artificial intelligence, we must first understand what in the end is intelligent? Smart, in fact, is the general term for intelligence and ability. World-renowned educational psychologist Howard Gardner proposed the famous "theory of multiple intelligences," he believes human individuals are independent of the existence of eight intelligences, are as follows:
- Visual - spatial intelligence, means sensitive to the line, shape, structure, color and spatial relationships and by the graphics plane and the three-dimensional shape out of their performance capability.
- Language - verbal intelligence, refers to reading and writing skills, the use of language to describe events, express ideas and ability to communicate with people.
- Communication - the exchange of intelligence, the ability to get along with people exchanges, the performance of perceived experience of others emotions, feelings and intentions and accordingly the ability to make an appropriate response.
- Knew - introspective intelligence, refers to the knowledge, insight and ability to reflect on their own, showing proper awareness and evaluate their emotions, motives, desires, personality, will, and my self-esteem form the basis of self-evaluation on the right of self-consciousness, the ability of self-discipline and self-control.
- Logical - mathematical Intelligence, refers to computing and reasoning ability, performance ability of the various relationships between things such as analogies, comparing the relationship, causality and logic, as well as sensitive thinking through mathematical and logical reasoning.
- Music - rhythm intelligent, referring to feelings, identification, memory, and the ability to alter the expression of the music, the performance capacity of individuals sensitive to music, including rhythm, tone, timbre and melody as well as by composing, playing and singing and other musical expression.
- Body - kinesthetic intelligence, the ability to use the limbs and trunk, showed better control of their own bodies, the events can make the appropriate physical reactions and body language to make good use of their ability to express thoughts and feelings.
- Naturalist intelligence, refers to the individual characteristics of the environment to identify and make use of the ability to classify and.
What is Artificial Intelligence
Artificial intelligence, that is, artificial intelligence, artificial out of the machine like humans think and act, so that the machine also has eight intelligences "theory of multiple intelligences" in the. Most people know about artificial intelligence mainly through science fiction, which the robot has a sense of human thought, emotion and exceptional ability.
But in reality artificial intelligence but with a far cry from science fiction, even disappointing, in reality, artificial intelligence can only recommend articles of interest to us, we can only help filter spam, can only be naive to chat with us only rigidly to help us translate, perhaps to complete autopilot in a simple environment. In reality artificial intelligence can only complete a single, relatively simple task, but not necessarily to complete the well, which is the gap between the ideal and the reality.
Weak AI
The ultimate goal of artificial intelligence is to be given to the machine consciousness thinking, thinking that it can work like the human brain. Overall, whether to have self-awareness and independent thinking for the sector, artificial intelligence can be divided into strong and weak AI Artificial Intelligence. Which refers to the ability of strong AI more human aspects have reached a level that can mimic human thought, awareness and ability to learn. And weak AI only focus on the completion of a particular task, certain aspects of human intelligence simulation, such as face recognition, voice recognition and so on.
Currently we often hear weak AI Artificial Intelligence in fact belong to the category, it can only solve the problem of a particular area, more is to act as a tool to use. Weak artificial intelligence based on big data and machine learning (including the current depth study over the fire), based on the model law is to learn things by a lot of data and algorithms calibration. Get a data model parameters by training and implementation decisions and forecast based on the model.
The strong artificial intelligence refers to the ability to have a variety of humans, such as independent thinking, self-awareness, emotions, inductive reasoning, and so on. For now, the strong field of artificial intelligence little substantive progress, do not have any engineering basic theory, more like a beautiful illusion.
History of artificial intelligence development
From artificial intelligence it was officially presented to the now sixty years later, during the development of artificial intelligence has experienced several times of prosperity and decline. At present, although good progress has been made, but the reality and the ideal gap is still large, the road ahead is still twists and turns.
In 1900, the International Congress of Mathematicians, mathematician Hilbert put forward the "mathematical problem of the future" where there are some problems associated with artificial intelligence. In fact, the incubation period of artificial intelligence dating back to BC, Aristotle's philosophy, he made the famous syllogism, deductive reasoning and even in terms of the impact so far; later mathematician Leibniz put forward in line with universal computing and reasoning, several mathematical logic laid the foundation; founded after the logician Boolean Boolean algebra, and for the first time with the symbol describes the basic inference rules.
In 1943, physicist Mike nerve Lodge and Pittsburgh built the first neural network model, MP model. In addition, the mathematician Alan Turing did a very important thing, is to design a Turing machine, which is the archetypal modern computers. And published "Computing Machinery and Intelligence" paper in 1950, this paper gives the definition of the machine and thinking, and to develop a "Turing test" standard, pass the test if the machine is considered intelligent.
1956 Dartmouth conference is known as the first year of artificial intelligence, but also mark the birth of artificial intelligence. This year, in the quiet town of Dartmouth College in the United States Luckysky, John McCarthy, father of artificial intelligence, artificial intelligence founder Marvin Minsky, Claude Shannon, founder of information theory, computer scientists Allen Newell, Nobel laureate Herbert Simon and other scientists came together to discuss how to mimic human intelligence machines. Meeting opened a full two months, although there is no general consensus reached, but it has a name: Artificial Intelligence.
After Dartmouth Artificial Intelligence Conference spurt began development in 1957, Rosenblatt invented the perceptron, 1959 Arthur Samuel scientists created a "machine learning" is a term, and gives the machine learning definition. In 1966 and 1972, respectively, they gave birth to the first chat robots and intelligent robots, and then with the decline in interest in artificial intelligence and funding dried up, in 1974, the first artificial intelligence began to enter the winter. Having gone through winter 1980 AI came back as an expert system, the expert system can provide decision-making capacity in specific areas. But soon after, in 1987, the depletion of artificial intelligence in the government and investors' money, entered the second winter.
In 1997, IBM's Deep Blue defeated world chess champion Garry Kasparov, becoming the first computer beat world chess champion. AI began in 2002 as a cleaning robot into the human family, until 2006, Facebook, Twitter, Netflix and other companies began to artificial intelligence technology into commercial systems. By the year 2011, IBM's Watson system has been able to in the quiz show in the human PK and win the championship.
The most important is that after the year 2006, the outbreak of large data and deep learning and development has been rapid, with two artificial intelligence implemented in some areas has been able to put on a par with humans. So in driving new round of technological wave, artificial intelligence continue to fall in many areas of applications, including face recognition, voice recognition, automatic pilot, precision marketing, personalized recommendations, intelligent customer service, security systems and so on.
Three Schools
Throughout the development of artificial intelligence, the researchers of different disciplines have different understanding of artificial intelligence, artificial intelligence, and therefore also had three schools. The traditional artificial intelligence school of thought is known as a symbol, symbolism is a major study of intelligent simulation method based on logical reasoning; while some people think that may be realized by neural network simulation of the brain that connect school of thought; in addition people think can find the answer from model organisms interact with the environment, the behavior is known as behaviorism.
Symbolism School
Symbols can be any certain school of thought patterns or symbols and transformed into some other operating system mode or symbols may produce intelligent behavior, it is committed to using the symbol of operating a computer to simulate human cognitive processes, and its essence is to simulate people abstract logical thinking brain, and human cognitive process described by some symbols, thereby realizing artificial intelligence. Symbolism focused on high-level behavior of human intelligence, such as reasoning, planning, knowledge representation and so on.
School of thought connected
Everyone's brain has a trillion neurons, they are intricately connected to each other, it is also considered a source of wisdom of mankind. So it is natural to think of the brain's ability to simulate intelligence by large number of neurons. Connecting school of thought considers the connection between the mechanisms and neural network learning algorithms and neural networks to produce intelligence.
Behaviorism
Behavioral school starting point and the other two schools is completely different, it is a perception based - Action intelligent simulation behavior. The school believes that individual behavior is a combination of a variety of body reaction to adapt to environmental change, its theoretical goal is to predict and control behavior.
review
This article explains what the whole artificial intelligence, artificial intelligence and the development of the three major schools of artificial intelligence, artificial intelligence to understand the situation of the discipline as a whole, and to know the current artificial intelligence is not science fiction inside AI gap between reality and the ideal is still very large.
Understanding of these cases artificial intelligence post, is not brought back your interest in artificial intelligence it? Next, the author will be ready "artificial intelligence principles of science" column, please pay attention!
Welcome concern: artificial intelligence, reading and feelings and talk about mathematics, distributed, machine learning, deep learning, natural language processing, algorithms and data structures, Java depth, Tomcat kernel and other related articles