Understand the past, present and future of artificial intelligence in one article

What exactly is artificial intelligence (AI)?

Artificial Intelligence , the English abbreviation is AI. It is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence.

The imitation of human intelligence processes by machines, especially computer systems, is called artificial intelligence. Expert systems, natural language processing, speech recognition and machine vision are several typical applications of artificial intelligence.

How does artificial intelligence work?

As enthusiasm around AI continues to grow, companies have been scrambling to show how their goods and services incorporate AI. What they call artificial intelligence is usually just one component of artificial intelligence, such as machine learning. AI requires specialized hardware and software to write and train machine learning algorithms. There is currently no single programming language that is synonymous with AI, but there are a few that stand out, including Python, R, and Java.

Artificial intelligence systems typically consume large amounts of labeled training data, evaluate the data for correlations and patterns, and then use these patterns to predict future states. By learning from millions of instances, a chatbot given text chat examples can learn to hold realistic conversations with humans. In contrast, image recognition programs can learn to recognize and describe items in photos.

Learning, reasoning and self-correction are the three cognitive functions that artificial intelligence programming focuses on.

Learning Process - This component of AI programming involves collecting data and developing rules to convert the data into usable information. These rules are called algorithms, and algorithms teach computer devices how to perform specific tasks step by step.

Inference Process - This area of ​​artificial intelligence programming is concerned with choosing the best method to achieve a given result.

Self-Correction Program – This feature of AI programming is designed to continually fine-tune algorithms and ensure they deliver the most accurate results.

Understand the various types of artificial intelligence classification

Since artificial intelligence research aims to make computers imitate human functions, the extent to which an artificial intelligence system can replicate human skills is used as a classification of artificial intelligence standard. Therefore, artificial intelligence can be classified into one of several categories based on how well the machine compares to humans in terms of diversity and performance.

In such systems, artificial intelligence that is able to perform more human-like functions and has a comparable level of competence is considered more advanced. In contrast, AI with limited functionality and performance is considered more straightforward and less evolved.

Based on this criterion, artificial intelligence is usually divided into two categories. One classification is based on the similarity of AI and AI-enabled robots to the human mind, and their ability to "think" and even "feel" like humans. According to this classification system, there are four categories of AI or AI-based systems: reactive machines, limited memory machines, theory of mind, and self-aware AI.

Reactive MachineHas no memory and is task-specific. Like Deep Blue, the IBM chess software that defeated Garry Kasparov in the 1990s. Deep Blue can identify pieces on the chessboard and make predictions, but due to its lack of memory, it cannot use past experiences to influence future experiences.

Limited Memory - Because these artificial intelligence systems have memory, they may use previous experiences to guide future judgments. This is how some of the decision-making mechanisms for self-driving cars are created.

Theory of Psychology is a word used in psychology. When applied to artificial intelligence, this means that machines have the social intelligence to understand emotions. This kind of artificial intelligence can predict human behavior and infer human intentions, which is an essential ability for artificial intelligence systems to become integral members of human teams.

Self-awareness - Artificial intelligence systems in this category have a sense of themselves, which gives them consciousness. Self-aware machines know their current situation. Currently, this form of artificial intelligence does not exist.

However, an alternative classification scheme more commonly used in technical terms is to classify technologies as: artificial narrow intelligence (ANI), artificial general intelligence (AGI), and artificial superintelligence (ASI).

Narrow Artificial Intelligence (ANI)

This form of artificial intelligence encompasses all existing artificial intelligence, including the most complex and capable ones built to date AI. Narrow artificial intelligence means that an artificial intelligence system can only complete one task independently while showing skills similar to humans. These machines can only accomplish what they were designed to do, giving them a limited or narrow range of capabilities. According to the above classification, these systems involve all types of artificial intelligence that are reactive and have limited memory. ANI even includes state-of-the-art artificial intelligence, which uses machine learning and deep learning to train itself.

Artificial General Intelligence (AGI)

The ability of an artificial intelligence agent to learn, perceive, understand, and function exactly like a human being is known as general artificial intelligence. By mimicking our multi-functional abilities, AI systems will have the same capabilities as humans. These systems will be able to independently build a large number of capabilities and make connections and generalizations across domains, significantly reducing training time.

Super Artificial Intelligence (ASI)

The birth of Artificial Super Intelligence (ASI) will undoubtedly mark the pinnacle of artificial intelligence research, because ASI will become the most Intelligent forms of competitiveness. In addition to imitating human intelligence, ASIs will be better in everything they perform due to vastly increased memory, faster data processing and analysis, and greater decision-making capabilities. Advances in AGI and ASI will lead to a scenario called the Singularity. While it's tempting to have such powerful tools at our disposal, these devices can jeopardize our existence, or at least our way of life.

What is the difference between machine learning and deep learning?

Machine Learning is a subset or application of artificial intelligence (AI) that allows systems to learn and grow from experience without having to be coded to that level. Machine learning uses data to learn and get the right results. Machine learning involves creating computer software that reads data and uses it to learn from itself.

Deep learning is a subset of machine learning, including artificial neural networks and recurrent neural networks. It uses algorithms and their methods to solve any complex problem. Algorithms are built in the same way as machine learning. However, there are many more layers of algorithms. The algorithm's network is called an artificial neural network. In simpler terms, it simulates the working mode of the human brain because all neural networks in the brain are connected. This is the concept of deep learning.

Relationship between the capabilities and data volume of simple statistical learning, traditional machine learning, and neural networks with various hidden layers

The following table compares machine learning to deep learning:

	Machine Learning	Deep Learning
1	Machine learning is a superset of deep learning	Deep learning Is a subset of machine learning
2	##Machine learning data is very different from deep learning data. Because it uses structured data.	The data format for deep learning is much different because it uses neural networks (ANN).
3	Machine learning is the next step in the development of artificial intelligence	Deep learning is the next step in the development of machine learning. Essentially, it refers to the depth of machine learning
4	Thousands are used in machine learning Data points	Millions of data points make up big data
5	Output: numerical value, such as score classification.	Anything from numbers to free-form features such as free text and sounds is acceptable.
6	Various automated algorithms are used to convert inputs into model functions and predict future actions.	To analyze data features and relationships, a neural network is used that sends data through processing layers.
7	Data analysts discover algorithms to evaluate certain variables in a data set.	Once implemented, algorithms essentially describe themselves in data analysis.
8	Machine learning is often used to stay ahead of the competition and learn new skills.	Deep learning is used to solve challenging machine learning problems.

https://www.researchgate.net/figure/Relationship-between-the -capabilities-and-amount-of-data-of-simple-statistical-learning_fig1_340134117.

Global Artificial Intelligence Market Scale

In 2021, the global artificial intelligence (AI) market was worth US$87.04 billion and is expected to reach US$1,597.1 billion in 2030 and 2022 The compound annual growth rate to 2030 is 38.1%. The global COVID-19 pandemic has been extraordinary and alarming, with higher-than-expected demand for this technology across all sectors compared to pre-pandemic levels. It is estimated that the global market will grow by 150% in 2022 compared to 2019.

##​

​https://www.php.cn/link/7439aebdba054f0d586d486ef2aff185​​

Growth Factors

Technological innovation has always been an important part of most industries. In recent years, the increasing popularity of digital technology and the Internet has greatly promoted the development of the global artificial intelligence industry. The huge expenditures on R&D by technology giants are constantly driving technological progress in various industries. Growing demand for AI technology across multiple end-use sectors such as automotive, healthcare, banking and finance, manufacturing, food and beverages, logistics, and retail is likely to drive the global AI market in the coming years.

The increasing popularity of numerous medical devices and the autonomous driving capabilities of new electric vehicles are significantly driving the growth of the global artificial intelligence market. The global digitalization trend is having a favorable impact on the market growth. The world's top IT giants, including Google, Microsoft, IBM, Amazon, and Apple, are increasing their efforts to promote and develop different artificial intelligence applications. Efforts by tech giants to improve access to artificial intelligence are expected to drive the growth of the global artificial intelligence market over the forecast period.

Over the past five years, medicine and healthcare have attracted the largest amount of private investment in AI globally ($28.9 billion).

The above is the detailed content of Understand the past, present and future of artificial intelligence in one article. For more information, please follow other related articles on the PHP Chinese website!