Computer vision, machine learning, natural language processing, robotics and speech recognition are the five core technologies of artificial intelligence, and all of them will become separate sub-industries.
Computer vision
Computer vision is the ability of a computer to recognize objects, scenes and activities from images. Computer vision technologies use sequences of image processing operations and other techniques to break image analysis tasks into manageable chunks. For example, some techniques are able to detect edges and textures of objects from an image, and classification techniques can be used to determine whether recognized features are representative of a class of objects known to the system.
Computer vision has a wide range of applications, including: medical imaging analytics used to improve disease prediction, diagnosis, and treatment; facial recognition used by Facebook to automatically recognize people in photos; used in security and surveillance to identify suspects; and in shopping, where consumers can now use their smartphones to photograph products for more purchasing options.
Machine vision, as a related discipline, refers broadly to vision applications in industrial automation. In these applications, computers recognize objects such as production parts in highly constrained factory environments, and thus have simpler goals than computer vision that seeks to operate in unconstrained environments. Computer vision is a work in progress, whereas machine vision is a "solved problem," a systems engineering topic rather than a research level topic. Some computer vision startups have attracted hundreds of millions of dollars in venture capital since 2011, as applications continue to expand.
Machine learning
Machine learning refers to the ability of a computer system to improve its performance by relying on data alone, without having to follow explicit program instructions. At its core, machine learning is the automatic discovery of patterns from data, which, once discovered, can be used to make predictions. For example, give a machine learning system a database of credit card transaction information such as transaction time, merchant, location, price, and whether the transaction is legitimate, and the system will learn patterns that can be used to predict credit card fraud. The more transaction data that is processed, the more accurate the predictions will be.
Machine learning has a wide range of applications, and it has the potential to improve the performance of almost everything for activities that generate huge amounts of data. In addition to fraud screening, these activities include sales forecasting, inventory management, oil and gas exploration, and public **** health. Machine learning techniques also play an important role in other areas of cognitive technology, such as computer vision, which improves its ability to recognize objects in massive amounts of images by continuously training and improving visual models.
Machine learning is now one of the hottest areas of research in cognitive technology, having attracted nearly $1 billion in venture capital investment in the 2011-2014 period alone. Google also spent $400 million in 2014 to acquire Deepmind, a company that researches machine learning technology.
Funding $400 million to acquire Deepmind, a company that researches machine learning technology.
Natural language processing
Natural language processing refers to the human-like text-processing capabilities that computers have. For example, extracting meaning from text, or even autonomously interpreting meaning from text that is readable, natural in style, and grammatically correct. A natural language processing system does not understand the way humans process text, but it can process it skillfully with great sophistication and sophistication. For example, automatically identifying all the people and places mentioned in a document; recognizing the core issues of a document; extracting terms and conditions from a human-readable contract and creating a table. These tasks are simply not possible with traditional text-processing software, which operates on simple text matches and patterns.
Natural language processing, like computer vision, incorporates a variety of techniques that help achieve its goals. Language models are built to predict the probability distribution of linguistic expressions, for example, the maximum likelihood that a given string of characters or words will express a particular semantic meaning. Selected features can be combined with certain elements of the text to recognize a piece of text, and by recognizing these elements it is possible to distinguish a certain type of text from others, e.g., spam from normal mail. Machine learning-driven categorization will be the criterion used to determine whether an email is spam or not.
Because context is so important to understanding the difference between "timeflies" and "fruitflies," the practical applications of natural language processing technology are relatively narrow. These include analyzing customer feedback about a particular product or service, automating the discovery of meaning in civil lawsuits or government investigations, and writing formulaic examples of corporate revenues and sports, to name a few.
Robotics
The integration of cognitive technologies such as machine vision and automated planning into extremely small but high-performance sensors, actuators, and cleverly designed hardware has given rise to a new generation of robots that have the ability to work alongside humans and have the flexibility to handle different tasks in a variety of unknown environments. Examples include drones and "cobots" that can share work with humans on the shop floor.
Speech recognition
Speech recognition focuses on technologies that automatically and accurately transcribe human speech. This technology has to deal with problems similar to those of natural language processing, with difficulties in handling different accents, background noise, distinguishing between homophones/anagrams ("buy" and "by" sound the same), as well as the need to have the ability to working speed to keep up with the normal speed of speech. Speech recognition systems use some of the same techniques as natural language processing systems, supplemented by other techniques such as acoustic modeling that describes sounds and the probability of their occurrence in a given sequence and language. Major applications of speech recognition include medical dictation, speech writing, voice control of computer systems, and telephone customer service. Domino_Pizza, for example, recently launched a mobile app that allows users to place orders by voice.
The industrialization of the above five technologies is an element of the industrialization of artificial intelligence. Artificial Intelligence will be a trillion-dollar market, or even a 10 trillion-dollar market, and will bring us some new and huge-capacity sub-industries, such as robotics, smart sensors, wearables, etc., of which the most anticipated is the robotics sub-industry.
There are many kinds of robot applications, which can be roughly divided into the following categories from the application level. The first category is industrial-grade robots, which are already well used by companies like Foxconn because labor costs are getting higher and higher and higher risks of employment, and robots can solve these problems. The second category is custodial robots, which can be used at home and in hospitals as caregivers for patients, the elderly or children, helping them to do things of a certain level of complexity. The demand for guardianship-grade robots in China is actually a little bit more urgent, because China's demographic dividend is declining and at the same time aging is on the rise, both of which are contradictions that robots can help solve. As a result, demand in this area accounts for a large share of the civilian market. The third category is adventure-grade robots, which are used for mining or exploring, etc., greatly avoiding the dangers that people have to go through. There are also military robots that are used to fight wars, etc.
The online media outlet Business Insider predicts that robots will replace humans in many positions: telemarketers, proofreaders, hand tailors, mathematicians, insurance underwriters, watch repairers, freight forwarders, tax preparers, image processors, bank account holders, librarians, typists, and so on. Because they are incredibly price-competitive. Research by the McKinsey Global Institute shows that when manufacturing wages in China increase by 10 to 20 percent a year, the price of robots globally is adjusted downward by 10 percent a year, with the cheapest low-ranking robot costing only half the average annual wage of an American. The international research organization Gueneng predicts that robots will lead to a new wave of global unemployment in 2020.
At the same time, the development of artificial intelligence technology will also allow many old industries to get a facelift type of new life, the most typical of which is the automobile industry. The automobile industry has existed for hundreds of years, during which the change is also very big, but driving a car is always a person, but in recent years, with Google and other companies invested heavily, the machine or some kind of automated system has been expected to replace the person to drive a car, thus forming a new industry with a huge market capacity, i.e., driverless car industry. The scale of this industry will also be trillions or even 10 trillion. Moreover, this industry will be superimposed and integrated with the new energy industry to form a composite industry of "Telematics + Telematics + Internet + Electric Vehicles" - in the future, we will use plug-in vehicles and hydrogen-fueled vehicles as power plants, thus making new energy vehicles part of the power grid. In the future, we will use plug-in cars and hydrogen-fueled cars as power plants, so that new energy vehicles will become part of the power grid and a supplier of new energy, just as some houses equipped with solar power systems are now suppliers of solar energy.
There's no doubt that, like the Internet, smart technology will penetrate almost all old industries. In a research report on the AI industry, Huatai Securities mentioned nine major industries: life service O2O, healthcare, retail, finance, digital marketing, agriculture, industry, commerce and online education. In fact, there are many more old industries that will get a new lease of life, such as the military, media, home, healthcare, life sciences, energy, public **** sector and even the virtual industry, which has been influenced by the development of VR/AR (Virtual Reality and Augmented Reality) technology. (Content from the machine people)
Wish to adopt, thank you