AI Concepts and Approaches
Alan Turing, a mathematician, made history a second time by asking a simple question: “Can machines really think?”
Turing’s paper “Computing Machinery and Intelligence” (1950), and its subsequent Turing Test, established the fundamental goal and vision of artificial intelligence. AI is the branch of computer science that seeks to answer Turing’s question. It’s the effort to duplicate or simulate human intelligence using machines. The expansive goal of artificial intelligence has given rise to many questions and debates. Because everyone has accepted no single definition of artificial intelligence.
AI can be defined as “building intelligent machines.” But it doesn’t explain what artificial intelligence is. What makes a machine intelligent, you ask? AI refers to the perception of their environment and then takes appropriate actions.
Four Types of Artificial Intelligence: Reactive Machines
Reactive machines follow the most fundamental AI principles. As their name suggests, they are capable of using their intelligence to perceive and respond to the world around them. Reactive machines cannot store memories and cannot use past experiences to guide decision-making in real-time.
Reactive machines are not designed to perform a wide range of tasks. They directly perceive the world. However, intentionally reducing a reactive machine’s worldview does not reduce costs. Instead, it means that this type of AI will be more reliable and trustworthy — it will respond the same way to every stimulus. An example of a reactive machine is the famous “Alternative Machine “Deep Blue” The chess-playing supercomputer, designed by IBM in the 1990s, defeated Gary Kasparov in a game. Deep Blue could only identify the pieces on a chessboard and determine each move’s moves based on the rules. It also acknowledged each piece’s position and determined the best move at the moment. The computer did not seek out potential moves from its opponent, nor was it trying to position its pieces better. Each turn was seen as an independent reality from any previous movement.
Although it is limited in scope and cannot be easily modified, reactive machine Artificial Intelligence can achieve a level of complexity and offer reliability when used to complete repeatable tasks.
Artificial intelligence with limited memory can store past data and make predictions. It’s useful for gathering information and making decisions. It also allows you to look back at the past to see any clues as to what might happen next. Reactive machines are less capable of storing limited memory artificial intelligence, which is more complicated and offers greater possibilities.
A team can continuously train a model to use new data or create an AI environment automatically renewed and trained. Using limited memory AI in machine learning requires six steps: First, the training data must be created to build the machine learning model. Finally, the model should be capable of making predictions. The model must also be capable of receiving feedback from humans or the environment. These steps must be repeated as a continuous cycle.
Three major machine learning models use limited memory artificial intelligence.
- Reinforcement learning: The program learns through trial and error to make better predictions.
- Long Term Short Term Memory (LSTM). Its past data is used to predict the next item in a series. While LTSMs consider more recent data the most important for making predictions, they also discount data from further back in time. However, they still use it to make conclusions
- Evolutionary Generative Adversarial Networks (E-GAN). The model evolves and explores new paths based on previous experiences. This model continuously seeks out a better path. It uses simulations and statistics (or chance) to predict future outcomes during its evolutionary mutation cycle.
THEORY OF MIND
Theory of Mind can only be described as theoretical. Unfortunately, we do not have the scientific and technological resources necessary to move on to the next level of artificial intelligence.
The psychological basis of the concept is that all living things have thoughts, emotions and that these affect one’s behavior. AI machines could then understand the emotions of humans, animals, or other machines and make decisions using self-reflection. They can then use that information to make their own decisions. The machines must process and comprehend the concept of “mind,”, emotions and other psychological concepts in real-time, creating a two-way relationship between artificial intelligence and humans.
The final step in AI’s journey to self-awareness will come after the Theory of Mind has been established in artificial intelligence. This artificial intelligence is human-level conscious and can recognize its existence and the emotional states of others. It would be able to understand the needs of others based on what they communicate and how they communicate it.
Artificial intelligence is based on the understanding of consciousness by humans and learning how to duplicate it so that self-awareness can be achieved.