Different types of artificial intelligence techniques

August 13, 2020  |  4 min read  |  0 views


Applications of artificial intelligence can be classified on the basis of type 1 and type 2 (depending on functionality).

Types  of  AI

Type 1 :(Based on Capabilities)

Type-1 has 3 types of artificial intelligence which is as follows.

Artificial Narrow Intelligence (ANI)

Artificial Narrow Intelligence is the most common form of Artificial Intelligence. Artificial Narrow Intelligence (ANI) systems are designed to solve a single problem and are capable of executing a single task well. Artificial Narrow Intelligence has strong capabilities, such as recommending a product to an e-commerce user or forecasting the weather.

It is the only Artificial Intelligence that still exists today. They are able to come close to human functioning in very specific contexts, and even surpass them in many instances, but excel only with limited parameters in very controlled environments.

Artificial General Intelligence (AGI)

Artificial General Intelligence (AGI) is a theoretical concept. It is defined as AI, which has human-level cognitive serve as in a wide variety of domains equivalent to language processing, image processing, computational functioning and logic and so forth.

We are still far from developing an AGI system. An AGI system would need to mimic human logic, encompassing thousands of artificial narrow intelligence systems.Even with the most advanced computing systems and infrastructure, such as Fujitsu or IBM’s Watson, it takes them 40 minutes to simulate one second of neuronal activity. It speaks to both the enormous complexity and interrelationship of the human brain, and the magnitude of the challenge of building AGI with our current resources.

Artificial Super Intelligence (ASI)

We are almost entering the science-fiction arena, but ASI is seen as the logical progression of AGI. An artificial super intelligence (ASI) system will be able to surpass all human capabilities.This includes things like rational decision making and better art making and emotional connections.Once we acquire Artificial General Intelligence, the AI ​​system will be able to rapidly improve its capabilities and move to those locations. Although the difference between AGI and ASI will be relatively small (some people say less as nanoseconds, because how fast it will learn artificial intelligence) the long journey ahead of us towards AGI makes it more like a concept in the future. She lives far away.

Type 2 :(Based on Functionalities)

Reactive Machines

AI has no prior memory of any kind and it never uses past information in the future. It is one of the oldest forms of AI, and together with it has limited capabilities. It does not have any memory based functionality. They also cannot learn and automatically react to limited inputs.

Example: Deep Blue, a chess-playing supercomputer at IBM of this types of AI, is a perfect example. It is famous for defeating International Grandmaster Gary Kasparov in the late 1990s. Deep Blue can identify individual pieces in chess and how each moves. It is able to beat its enemies all around. Depending on the option, it selects the best possible move. However, it does not have the ability to learn from previous tricks as these machines have no memory of their own.

Limited Memory

This type of AI has the capable of reactive machines having some memory capacity, so that they can use past data / experience to make better experience for future events. Most commonly applications around us also fall into this category. These AI applications can be trained by a large amount of training data that they store in their memory as a reference model.

Example: Limited memory technology is used for repetitive tasks like self-driving cars. They store the nature / size of barriers in a hundred other types of data such as GPS location such as the speed of nearby cars, drive like a human.

Theory of mind

This type of AI is just a concept and will require some improvement before it can work. It is currently being researched and will be used to better understand people’s feelings, needs, beliefs and thoughts. Artificial Emotional Intelligence is a budding industry and repository of interest, but to reach this level, further effort will be required over time. To truly understand human needs, the AI ​​machine has to be seen as individuals whose minds are shaped by many factors.

Example: According to researcher Winston, he showed in his research a prototype of the robot, which can walk down the small corridor with other robots coming from the opposite direction, AI has the ability to intercept and isolate other robots and AI can Can bend right, left or any other way so that there is no collision with the robot coming from the front. According to Wilson, this robot determines its action based on “common sense” of how other robots will move.

Self-awareness

This is the final stage of AI. It is only fictional in current existence and can only be found in science fiction films. This type of AI can not only understand and provoke human emotions, but can also evoke feelings of its own. Such AI is decades if not centuries away from materiality. This is because once it becomes self-aware, AI can enter self-preservation mode, treating humanity as a potential threat and attempting directly or indirectly to end humanity is.