What is AI?
Artificial Intelligence by direct translation is the artificial ability to infer, understand, and learn knowledge to solve problems. This artificial feature aims to copy the natural way humans comprehend information and use knowledge to make decisions. In technical terms, Artificial Intelligence is a branch of computing that enables computer systems or machines to simulate (mimic) human intelligence processes such as reasoning, learning, solving problems, and making decisions. How is AI able to mimic human intelligence?
Understanding human Intelligence
Human Intelligence is a mental process that enables the brain to reason, learn, and solve problems. In addition, Intelligence fuses cognitive functions such as perception, attention, and memory. Different regions of the brain are responsible for various activities. The amygdala creates a sensation of fear, the hippocampus encodes memories, and the prefrontal cortex is responsible for functioning and behaviors. However, Intelligence is enabled through a network of billions of neurons across several brain regions. Neurons are the information messengers that use signals to send information between different areas of the brain and the entire body. When a neuron's incoming signal is input, the signals add up until a particular threshold. Once they exceed the threshold, the neuron is triggered to send an outgoing impulse. They begin forming as we develop in the womb, where a neuron is connected to thousands of other neurons, forming a complex communication network.
Artificial Intelligence mimics the mental process through a computerized process that can reason, learn, and solve problems. The fundamental conception of AI begins with machine learning (ML) algorithms, which are commands followed by a computerized machine. Secondly, large data sets are used by a machine as a learning reference. Thirdly, Natural Language Processing (NLP) enables the machine to interpret, understand, and generate text and speech in human language. For instance, Siri and Zuri use NLP. Lastly, neural networks are interconnected nodes; the nodes (made up of millions of artificial neurons) process and learn from data to solve a problem. Similar to the human brain, each node connects to others with its associated threshold value that, when exceeded, sends data to the next node. For the neural networks to function independently, they are trained using the data they are supposed to learn from. The more reference data they have, the better they learn. The more trained they are, the more accurately they will reason and solve problems.
Now that is the artificial human Intelligence. Let's give tech a standing ovation.
Despite the various components of an AI, it is highly characterized by neural networks. The idea of neural networks was first introduced in 1936 by Alan Turing, an English Mathematician when he published 'On Computable Numbers.' This publication was the most famous theoretical paper in the history of computing, which mathematically described an imaginary computing device capable of implementing any computer algorithm. It was coined the Turing machine. In 1943, American neurophysiologist and cybernetician Warren McCulloch and a self-taught Logician Warren Pitts built on the Alan Turin idea and published the first mathematical model of a neural network called 'A Logical Calculus of the Ideas Imminent in Nervous Activity.' McCulloch and Pitss's paper described brain functions and how connected neural networks can have immense computational power.
In 1955, the concept was initialized by Allan Newell, a cognitive psychologist; John Clifford, a systems programmer; and Herbert A. Simon, a political scientist, economist, and sociologist. The trio developed the Logic Theorist. The Logic Theorist was the first program designed to mimic the problem-solving skills of a human being. The program was presented at the Dartmouth Summer Research Project on Artificial Intelligence hosted by Marvin Minsky and John McCarthy in 1956. After the conference, McCarthy brought together great researchers from various fields to discuss Artificial Intelligence. Despite the failure to agree on standard methods for the AI field, everyone unanimously agreed that AI was achievable. The event significantly catalyzed the years of AI research.
