Artificial Intelligence Chapter 1 Part 2 Artificial Intelligence

Artificial Intelligence Chapter 1 - Part 2 Artificial Intelligence (605451) Dr. Hassan Al-Tarawneh

What is Artificial Intelligence? • Artificial Intelligence (AI) is a branch of computer science devoted to creating computer software and hardware to attempt to mimic human intelligence or human intelligent behaviour

What is Intelligent Behavior? • Learn from experience – – Respond flexibly in situation that are New or not anticipated Ambiguous (not clearly defined) Specified by contradictory information • Apply ‘learnt’ or ‘programmed’ knowledge to solve problems • Use reasoning in solving problem • Understand infer facts and rules • Recognize the relative importance of different elements in a situation

What is AI concerned with? • Mainly AI is concerned with two basic ideas: 1. It involves studying the thought processes of humans to understand what intelligence is. 2. It deals with representing thought processes using computers to create artificially intelligent entities for testing theories of intelligence.

Some Definition of AI • AI is a term that compasses many definitions: – “The goal of work in AI is to build machines that performs tasks normally requiring human intelligence” – N. Nilsson – “The goals of AI research is to construct computer programs which exhibit behaviors that we call intelligent behavior when we observe it in human beings – E. Feigenbaum & J. Feldman – “AI is the science of making computers smart. It is practiced both by those who want to make computers more useful and by those who want to understand the nature of intelligence” – Patrick H. Winston

Some Definition of AI • “AI is the ability of machines to do things that people would require intelligence. • AI research is an attempt to discover and describe aspects of human intelligence that can be simulated by machines. • Alternatively, AI research may be viewed as an attempt to develop an mathematical theory to describe the abilities and actions of things (natural or man-made) exhibiting ‘intelligent’ behavior and server as a calculus for the design of intelligent machines“ – Philip C. Jackson

Artificial Intelligence vs Natural Intelligence AI Advantages AI offers ease of duplication and dissemination of knowledge Natural Intelligence Disadvantage Knowledge duplication and transfer of expertise requires apprenticeship AI offers more permanent storage of Natural Intelligence is perishable – knowledge people forget or be inaccessible AI being a computer technology is consistent, thorough and formal Natural Intelligence can be erratic as that is the way most people are AI can be documented – decisions made by the system can be easily introspected and documented Natural Intelligence is difficult to document - people may do things but can not explain how they did it AI, from a commercial point of view, Experts having knowledge of a is less expensive - a given system certain domain are more expensive may be bought only once and used frequently

Natural Intelligence vs Artificial Intelligence Natural Intelligence Advantages Natural Intelligence is creative – the ability to acquire knowledge is inherent in humans Natural intelligence work with sensory experience Humans have the natural ability to reason, which is applicable to a wide context of experience AI Disadvantage AI systems have a static knowledge, they work with only the programmed knowledge AI system can work with only symbolic input AI system have a very limited scope and can reason to a limited extend based on the reasoning mechanism implemented

Comparing Humans and Computers • Computer are good at – Processing large amount of data – Storing (permanently) and retrieve large amount of data – Working at extremely fast speeds – Doing extensive arithmetic calculation with great accuracy – Working for long periods without fatigue

Comparing Humans and Computers • Humans are good at – – – – Recognizing various things, either seen before or not Recognizing the relationship patterns between things Thinking – common sense reasoning Combining all types of sensory inputs Acting appropriately in novel situations Learning new things and changing behavior patterns Making decisions even when given incomplete information – Working with noisy, incomplete or ambiguous information

Comparing Humans and Computers • The tasks humans do naturally however are extremely difficult for a computer program to accomplish • If computers are to be termed as intelligent, they must be able to do the same kind of task as humans do naturally

AI in Practice • AI is an combination of many different sciences and technologies: • Linguistics – Computational linguistics – Phycholinguistics – Socio - linguistics • Phychology – Cognitive science – Phycholinguistics • Philosophy – Logic – Philosophy of mind – Philosophy of language • Electrical Engineering – – – Image processing Speech processing Pattern recognition Robotics Machine learning Neural networks • Optimization – Scheduling • Management – Management information system – Decision making

AI in Practice • AI provides the framework and methodology for several growing commercial technologies • Expert System – Computerized advisory and/or decision making systems that attempt to imitate the reasoning process and knowledge of experts in solving domain-specific problems • Natural Language Processing – Natural language technology attempts to give computer users the ability to communicate with computer in natural language. Also deals with understanding textual language for various applications • Speech recognition – The recognition of spoken language by a computer – breaking up of speech into meaningful units, i. e. words

AI in Practice • Robotics and Sensory Systems – Combining robots having mechanical motions with sensory system to produce machines of varying intelligence and abilities. Here a robot senses its environment and modifies its behavior on the basis of the information gained. • Computer vision and scene recognition – To understand digitized visual information received from a sensor and then making a decision – interpreting a visual scene • Intelligent Computer-Aides Instruction – To develop machines that can tutor humans in the interactive manner. Such systems have a vast knowledge base and are used for teaching problems solving skills, simulations, discovery and learning, drill and proactive and so on.

AI in Practice • Game playing – To develop intelligent programs that can play games such as chess • Neural network – Systems that attempt to mimic the neural structure of the human brain. Information processing is carried out in a manner close to that of the human brain • Intelligent Databases – data mining – Intelligently managing large databases, facilitating the furnishing of complex queries and automatically gleaning hidden information from databases – data mining

AI in Practice • Exemplar Systems – – – – Language translation systems Speech to text system Air traffic control systems Automates personal systems Supervisory systems Intelligent highways – traffic monitoring Robots for hazardous conditions Expert system for law, medicine Neural network based forecasting – finance, stocks, medicine Executive summary producing systems Automatic programming Summarizing news from papers Intelligent design – architecture, mechanical and electrical systems Chess playing systems – deep blue Medical diagnostic systems

AI in Practice • AI is a subject of scientific investigation, with the following working practices : – Researchers construct theories about what AI programs are capable of and test them with mathematical analysis or experiments – This may involve decision making and problem solving using symbolic processing. A symbol is a string of characters that stands for some real-world concept