History of Artificial Intelligence PDF

Ministry of Higher Education and Scientific Research ‫وزارة التعليم العالي والبحث العلمي‬ University of Continuous Education...

Ministry of Higher Education and Scientific Research ‫وزارة التعليم العالي والبحث العلمي‬ University of Continuous Education ‫جامعة التكوين المتواصل‬ Master's Title: Organizational Management Artificial Intelligence - Chapter 1 History of Artificial Intelligence Lessons prepared by Dr.: Benbouras Mohammed Amin Lecturer Class "A" at the Faculty of Civil Engineering, USTHB 1- Introduction: It is no secret to any student in the field of computer science that there is a growing interest in artificial intelligence (AI) and machine learning (ML). The same goes for the average technology user, who is witnessing significant advancements in the machines and technologies surrounding them. From the voice-activated assistant in their phone to their car and other smart devices, all the way to the military sector, where most mechanisms and aircraft now rely on AI and ML technologies, the competition has grown fierce among advanced nations. This prompted Russian President Vladimir Putin to declare, "Whoever leads in AI will rule the world." On the other hand, inventions and discoveries are emerging every day, and major companies worldwide are competing to see who can make the most significant progress and steal the spotlight. One example is Neuralink's recent announcement that it has made strides toward implanting chips in the brain to enable direct communication between the machine and the brain. This development in AI and ML didn’t come about by chance but is the result of great efforts and work throughout history. In these lessons, we will attempt to provide an overview and introduction to AI and ML. 2- The History of Artificial Intelligence: Although AI is less than 100 years old, it has a rich history stretching back to the 1940s. Below is a review of notable achievements in AI by decade:  1943: The first model based on artificial neurons was proposed, where each neuron could be either active or inactive depending on the signals it received from neighboring connected neurons.  1948: The first two AI-programmed robots, Elmer and Elsie, were created. These were self-controlled robots capable of navigating obstacles using light and touch.  1950: The Turing Test was introduced, which measures a machine's ability to exhibit intelligent behavior indistinguishable from that of a human. Core AI concepts like machine learning, genetic algorithms, and reinforcement learning also emerged during this time.  1951: The Stochastic Neural Analog Reinforcement Computer (SNARC) was designed as the first computer to work with neural networks.  1958: Lisp, a programming language specifically designed for AI, was developed. A paper was also published on the Hypothetical Advice Taker, an AI system capable of learning from experience.  1964: ELIZA, the first natural language processing program, was introduced. It became the foundation for all modern chatbots.  1974-1980: This period is known as the first AI winter, during which funding for AI projects decreased due to limited progress, and AI’s impact on daily life applications declined. One of the criticisms AI faced during this time was its inability to solve the combinatorial explosion problem, limiting its applicability to small problems and datasets.  1980: The first successful commercial expert system designed to simulate human decision-making capabilities was launched.  1987-1993: This period is known as the second AI winter, as early AI systems were rule-based, which restricted their applicability and made them unable to solve real-world, human problems.  1997: AI made its first major win in the world of chess, as the Deep Blue supercomputer, programmed with artificial neural networks, defeated world chess champion Garry Kasparov, who had been undefeated in seven consecutive world championships (from 1984 to 1997). This victory sparked what is known as the AI Spring (the AI boom), where AI rapidly spread across many fields.  2005: Stanford University developed the self-driving car Stanley, which won the self-driving car challenge. The U.S. military also began investing in autonomous robots.  2009: Graphics Processing Units (GPUs) were used for the first time to train deep learning neural networks. The use of specialized hardware accelerated the training of complex networks on vast amounts of data, ushering in a new era of deep learning and AI.  2011: IBM’s Watson, a question-answering system, defeated the best players in the world on the American quiz show Jeopardy!, successfully understanding and answering questions. This marked a breakthrough in AI’s ability to understand natural language.  2012: An AI system was introduced that could instantly translate spoken English into spoken Chinese.  2021: Fully autonomous driving systems emerged, utilizing neural networks trained on the behavior of hundreds of thousands of drivers.  2022: The chatbot Generative Pre-Trained Transformer (ChatGPT) appeared, based on large language models. These models are finely tuned using both supervised and reinforcement learning techniques to simulate human conversations.