What is artificial intelligence?

Artificial intelligence (AI) is a broad branch of computer science related to the construction of smart machines capable of performing tasks that typically require human intelligence.

What Are The Four Types of Artificial Intelligence?

  • Reaction machines
  • Limited memory
  • Theory of mind
  • Self-awareness

What are the Examples of Artificial Intelligence?

  • Siri, Alexa, and other smart assistants
  • Self-driving cars
  • Robo Advisor
  • Chatbots
  • Email spam filters
  • Netflix Recommendations

How does Artificial Intelligence Work?

AI Perspectives and Concepts

Less than a decade after the Nazi encryption machine broke the enigma and helped Allied forces win World War II, mathematician Alan Turing changed history for the second time with a simple question: “Can machines think? ”

Turing’s dissertation “Computing Machinery and Intelligence” (1950) and subsequent Turing tests established the basic purpose and vision of artificial intelligence.

At its core, AI is the branch of computer science that aims to answer questions in the affirmative sense. It is an attempt to mimic or mimic human intelligence in machines.

The broader purpose of artificial intelligence has raised many questions and debates.

Can Machines Think? – Alan Turing, 1950

One of the major limitations of describing AI as mere “intelligent machines” is that it does not really explain what artificial intelligence is. What makes the machine intelligent? AI is an interdisciplinary science in many ways, but advances in machine learning and in-depth learning are creating an ideal transformation in almost every sector of the tech industry.

Artificial Intelligence: A Modern Approach, authors Stuart Russell and Peter Norvig, in their important textbooks, approached the question by combining their work around the topic of intelligent agents in machines. With that in mind, AI is the study of agents that receive feedback from the environment and perform actions.

View all AI Companies

Norvig and Russell proceed to explore four different methods that have historically defined the field of AI:

  1. – Human thinking
  2. Think rationally
  3. Working humanely
  4. Act rationally

The first two ideas deal with the process of thinking and reasoning, while the second deals with behavior. Norvig and Russell specifically focus on rational agents who work to achieve the best results, noting that all the skills required for the Turing Test allow the agent to work rationally as well. (Russell and Norrug 4).

Patrick Winston, Ford Professor of Artificial Intelligence and Computer Science at MIT, praised AI, saying that algorithms, enabled by limitations, emerge through representations on the loops that bind thinking, perception and action together. Support targeted models.

While these definitions may seem abstract to the average person, they help focus the field as a field of computer science and provide a blueprint for machines and programs, along with machine learning and other subsets of artificial intelligence. Provide

Four Types of Artificial Intelligence

Reactive Machines

A reaction machine follows the many basic & common principles of AI and, as the name implies only its intelligence is able to and react to the world in front of it. A reactive machine cannot store memory and as a result, cannot rely on past experiences to report real-time decision-making.

Understanding the world directly means that only a limited number of reaction machines are designed to perform specific tasks. However, deliberately narrowing the global view of a reaction machine is not a cost-cutting measure in any way, and instead means that this type of AI will be more reliable and dependable. It will react similarly to the same stimulus each time.

A well-known example of a reaction machine is Deep Blue, which IBM designed as a chess-playing supercomputer in the 1990s and defeated international grandmaster Gary Kasparov in a game. Deep Blue was only able to identify the pieces on the chessboard and to know how each movement moves based on the principles of chess, recognizing the current position of each piece, and the most logical at the moment. What will happen? The computer was not pursuing possible future moves by its adversary or trying to keep its pieces in a better position. Each turn was seen as its own reality, separate from any other movement that had already taken place.

Another example of a game-playing reactive machine is Google’s Alfano. AlphaGo is also unable to predict future moves, but relies on its neural network to monitor current game progress, giving it a lead over Deep Blue in more complex games. AlphaGo also beat world-class rivals in 2016 by defeating champion Go player Lee Sedol.

Although limited in scope and not easily modified, the reactive machine can achieve artificial intelligence levels of complexity and offers reliability when designed to perform reliable tasks.

Limited Memory

Limited memory Artificial intelligence has the ability to store previous data and predictions when information is gathered and possible decisions are weighed – primarily to find out what happened in the past. Maybe. Limited memory Artificial intelligence is more complex and offers more possibilities than reaction machines.

Limited memory AI is formed when a team constantly trains a model on how to analyze and use new data or create an AI environment so that models can be automatically trained and updated. When using limited memory AI in machine learning, six steps must be followed: training data must be created, a machine learning model must be created, a model must be able to make predictions, a model must be able to obtain human or environmental impressions. Should be, this feedback should be stored as data, and these steps should be repeated as a cycle.

There are three major models of machine learning that use limited memory artificial intelligence:

  • Learning to help, who learns to make better predictions through repeated trials and errors.
  • Long Short Term Memory (LSTM), uses past data to help predict the next item in a sequence. LTSMs see recent information as the most important when making further predictions in the past and discounting data, although they still use it to draw conclusions.
  • Evolutionary Generative Advertising Networks (E-GAN) which evolve over time, grow with each new decision to find slightly different paths based on past experiences. This model is constantly looking for a better way and uses simulations and statistics or predictions to predict the outcome during its evolutionary mutation cycle.

Theory of Mind

That’s the theory of mind – theoretical. We have not yet acquired the technical and scientific skills necessary to reach this next level of artificial intelligence.

This concept is based on the psychological basis of understanding that other living beings have thoughts and emotions that influence one’s behavior. In terms of AI machines, this would mean that AI could understand how humans, animals, and other machines feel and make decisions through self-reflection and determination, and then use that information to make its own decisions. Will Basically, machines need to be able to understand and act on the concept of “mind”, a fluctuation between the fluctuations of emotions in decision making and other Latin concepts in real-time, between people and artificial intelligence. To create a relationship

Self-Awareness

Once the Theory of Mind is established in artificial intelligence, sometime in the future, the final step will be to become self-aware of AI. This type of artificial intelligence is aware of the human level and understands its own existence in the world as well as the presence and emotional state of others. Be able to understand what others may need based on not only what they talk about but also how they talk.

Self-awareness in artificial intelligence relies on human researchers who understand the basis of consciousness and then learn how to replicate it so that it can be built into machines.

How is AI Used?

Addressing a crowd at the 2017 Japan AI Experience, Jeremy Achin, CEO of DataRobot, began his speech with the following definition of how AI is used today

AI is a computer system capable of performing tasks that typically require human intelligence … Many of these artificial intelligence systems run on machine learning, some of them through deep learning. Walk and some of them are reinforced by very boring things like rules.

  • Artificial intelligence has two types:Narrow AI: Sometimes called “weak AI”, this type of artificial intelligence operates within a limited context and mimics human intelligence. Tung AI often focuses on performing a single task very well and although these machines may seem intelligent, they are operating under more obstacles and limitations than basic human intelligence.
  • Artificial General Intelligence (AGI): AGI, sometimes called Strong AI, is the type of artificial intelligence we see in movies, such as robots from the West World or data from Star Trek: The Next Generation. AGI is a machine that has general intelligence and like human beings, it can use this intelligence to solve any problem.

Narrow Artificial Intelligence

Narrow AI is all around us and is the most successful perception of artificial intelligence to date. With its focus on performing specific tasks, Narrow AI has achieved a number of successes over the past decade that have “achieved significant social benefits and contributed to the nation’s economic strength”, “artificial intelligence According to “Preparing for the future”. The 2016 report was released by the Obama administration.

Some examples of narrow AI include:

  • Google Search
  • Image recognition software
  • Siri, Alexa, and other personal assistants
  • Self-driving cars
  • IBM’s Watson

Machine Learning and Deep Learning

Much of Narrow AI is powered by successes in machine learning and deep learning. Understanding the difference between machine learning & artificial intelligence deep learning can be confusing. Venture capitalist Frank Chen offers a good overview of the differences between them, noting

Simply put, machine learning feeds computer data and uses statistical techniques to help it “learn” that eliminates the need for millions of lines of written code, without being programmed specifically for a task. how to put it together for use with the business. Machine learning consists of both supervised learning (using labeled datasets) and non-supervised learning (using unlabelled datasets).

Deep learning is a type of machine learning that drives input through biologically impaired neural network architecture. Neural networks have many hidden layers through which data is processed, allowing the machine to go “deeper” in its learning, make connections, and weigh the input for best results.

Artificial General Intelligence

The creation of a machine with human-level intelligence that can be applied to any task is a hollow grill for many AI researchers, but the search for AGI is fraught with difficulties.

time has not reduced the difficulty of building a machine with a full set of basic cognitive abilities but The search for a universal algorithm for learning and practicing in any environment is nothing new AGI has long been a muse of dystopian science fiction, in which highly intelligent robots subdue humanity, but experts agree that this is not something we should be worried about any time soon. Required.

A computer science graduate. Interested in emerging technological wonders that are making mankind more approachable to explore the universe. I truly believe that blockchain advancements will bring long-lasting revolutions in people’s lives. Being a blogger, I occasionally share my point of views regarding the user experience of digital products.

Leave a Comment