![16
Rational agents
• An agent is an entity that perceives and
acts
• This course is about designing rational
agents
• an agent is a function from percept
histories to actions:
[ f : P* → A]
• For any given class of environments and
tasks, we seek the agent (or class of
agents) with the best performance
• Note: Computational limitations make
perfect rationality unachievable
→ Design the best program for given
machine resources](https://image.slidesharecdn.com/ch1introductiontoaiapplications-241018112832-60f2bbd7/85/Ch-1-Introduction-to-AI-Applications-pdf-16-320.jpg)
Ch 1 Introduction to AI Applications.pdf
- 1. 1 ARTIFICIAL INTELLIGENCE Introduction: Chapter 1 اإلصطناعي الذكاء
- 2. How would you define “intelligence”? What is the common definition of “AI”? What are the AI sub-topics? Which topics failed? successful? Why? Do you know any AI real application? Should artificial intelligence simulate natural intelligence? What is the relation between AI and logic? Do you think that computers or machines will ever be as intelligent as humans? What is the main advantage of computers over people and vice versa? How far is AI from reaching human-level intelligence? When will it happen. Are computers fast enough to be intelligent? Do you definitely agree that AI augmenting human capability and capacity, or it will damage the human life?
- 4. 2021 2005
- 5. 5 What is Intelligent There are many definitions of intelligence. A person that learns fast or one that has a vast amount of experience, could be called "intelligent". However for our purposes the most useful definition is: systems comparative level of performance in reaching its objectives persons are not intelligent in all areas of knowledge, they are only intelligent in those areas where they had experiences.
- 6. 6 AI Goals • Artificial Intelligent is the part of computer science with designing intelligent computer systems, that is, systems that have characteristics associate with intelligence in human behaviour – understanding language, learning, reasoning, solving problems……………… • Scientific Goal To determine which ideas about knowledge representation, learning, rule systems, search, and so on, explain various sorts of real intelligence. • Engineering Goal To solve real world problems using AI techniques such as.. knowledge representation, learning, rule systems, search, and so on.
- 7. Artificial Intelligence in the Movies
- 8. Why the interest in AI? Search engines Labor Science Medicine/ Diagnosis Appliances What else?
- 9. 9
- 10. 10 What is AI? Views of AI fall into four categories: • Thinking humanly: systems that thinks like humans, (machine with mind). Activities as decision-making, problem solving, learning,…… • Thinking rationally: the study of thinking faculties. • Acting humanly: systems that acting like humans, the study of how to make computers do things. • Acting rationally: The study of designing intelligent agents The textbook advocates “Acting Rationally"
- 11. 11 Systems that think like humans Systems that think rationally Systems that act like humans Systems that act rationally Turing test Cognitive science Logic Agents
- 12. 12 How do Humans do Intelligent Things? • It seems natural to try to base our AI systems on the human nervous system. This can be broken down into three stages that may be represented in block diagram form as: Receptors collect information from the environment, and effectors generate interactions with the environment. The flow of information between them is represented by arrows – both forward and backward. What we generally describe as “intelligence” is normally carried out in the central stage – in the brain. The brain is known to consist of an interconnected network of neurons, and the study of neural networks is now a major sub-field of AI.
- 14. 14 Acting Humanly: Turing Test • Turing (1950) "Computing machinery and intelligence": • "Can machines think?" → "Can machines behave intelligently?" • Operational test for intelligent behavior: the Imitation Game • Alan Turing's discussed the conditions for considering a machine to be intelligent. He discusses that if the machine could successfully pretend to be human to a knowledgeable observer then you certainly should consider it intelligent. • Ask questions of two entities, and receive answers from both • If you can’t tell which of the entities is human and which is a computer program, then we should therefore consider the computer to be intelligent
- 15. 15 Sub-fields of Artificial Intelligence AI now consists many sub-fields, using a variety of techniques, such as: Neural Networks – e.g. brain modeling, time series prediction, classification Evolutionary Computation – e.g. genetic algorithms, genetic programming Computer Vision – e.g. object recognition, image understanding Robotics – e.g. intelligent control, autonomous exploration Expert Systems – e.g. decision support systems, teaching systems Speech Processing– e.g. speech recognition and production Natural Language Processing – e.g. machine translation Machine Learning – e.g. decision tree learning, version space learning Most of these have both engineering and scientific aspects.
- 16. 16 Rational agents • An agent is an entity that perceives and acts • This course is about designing rational agents • an agent is a function from percept histories to actions: [ f : P* → A] • For any given class of environments and tasks, we seek the agent (or class of agents) with the best performance • Note: Computational limitations make perfect rationality unachievable → Design the best program for given machine resources
- 17. 17 Examples of AI Agents Humans Programs Robots___ senses keyboard, mouse, dataset cameras, pads body parts monitor, speakers, files motors, limbs Ch2 Intelligent Agents (input, output, Types, ……)
- 18. AI Complex?
- 21. 21 The Roots of AI AI has roots in a number of older sciences , particularly: • Philosophy • Logic/Mathematics • Computation • Psychology/Cognitive Science • Biology/Neuroscience • Evolution • By looking at each of these in turn, we can gain a better understanding of their role in AI, and how these underlying the developed to play that role.
- 22. 22 History of AI: 1952- 1969 • Great successes! – Solving hard math problems – game playing – LISP was invented by McCarthy (1958) – McCarthy went to MIT and Marvin Minsky started lab at Stanford (Both powerhouses in AI to this day) History of AI: 1966 - 1973 • Reality – Systems fail to play chess and translate Russian – neural networks was exposed (neural networks did not return to appear until late 1980s)
- 23. 23 AI History: 1969 - 1979 • Knowledge-based Systems (Expert systems) – Problem: General logical algorithms could not be applied to realistic problems – Solution: accumulate specific logical algorithms • DENDRAL – infer chemical structure • AI History: 1987 -2000 • AI becomes a science – More repeatability of experiments – More development • Intelligent Agents (1994) – AI systems exist in real environments with real sensory inputs
- 24. 24 2000- Where are We Now? – Autonomous planning: scheduling operations aboard a robot – Game playing: Kasparov lost to IBM’s Big Blue in chess – Autonomous Control: CMU’s NAVLAB drove from Pittsburgh to San Francisco under computer control 98% of time – Stanford vehicle wins 2006 DARPA Grand Challenge CMU’s 2005 vehicle falls crashes at starting line – Logistics: organized the time tables for any task. – Robotics: remote heart operations. – human genome, protein folding, drug discovery. – stock market …………………….etc.
- 25. 1. AI Application in E-Commerce Recommendation engines , chatbots help improve the user experience , Credit card fraud and fake reviews. 2. Applications Of Artificial Intelligence in Education Creating Smart Content, Voice Assistants, Personalized Learning 3. Applications of Artificial Intelligence in Lifestyle Autonomous Vehicles, Spam Filters, Facial Recognition, Recommendation System 4. Applications of Artificial Intelligence in Navigation GPS technology can provide users with accurate, timely, and detailed information to improve safety. 5. Applications of Artificial Intelligence in Robotics : Carrying goods in hospitals, factories, and warehouses, Cleaning offices and large equipment 25 6. Applications of Artificial Intelligence in Human Resource job candidates' profiles 7. Applications of Artificial Intelligence in Healthcare detect diseases and identify cancer cells, analyze chronic conditions , early diagnosis. 8. Applications of Artificial Intelligence in Agriculture computer vision, robotics 9. Applications of Artificial Intelligence in Automobiles self-driving . 10. Applications of Artificial Intelligence in Social Media determine what posts you are shown. DeepText can understand conversations better. 11. Applications of Artificial Intelligence in Marketing • Data mining 2000- Where are We Now?
- 26. 26
- 27. 27
- 28. 28
- 29. 29
- 30. 30
- 31. 3 1 Genetic Algorithms • Basic scheme – (1)Initialize population – (2)evaluate fitness of each member – (3)Selection of the best Chromosomes – (4) Crossover – (5) introduce random mutations in new generation – Continue (2)-(3)-(4) until prespecified number of generations are complete • Start with k randomly generated states (population) • Evaluation function (fitness function). Higher values for better states. • Produce the next generation of states by selection, crossover, and mutation
- 32. A Simple Example The Traveling Salesman Problem: Find a tour of a given set of cities so that – each city is visited only once – the total distance traveled is minimized Representation Representation is an ordered list of city numbers known as an order-based GA. 1) London 3) Dunedin 5) Beijing 7) Tokyo 2) Venice 4) Singapore 6) Phoenix 8) Victoria CityList1 (3 5 7 2 1 6 4 8) CityList2 (2 5 7 6 8 1 3 4)
- 33. TSP Example: 30 Cities 0 20 40 60 80 100 120 0 10 20 30 40 50 60 70 80 90 100 y x
- 34. Solution i (Distance = 941) 0 20 40 60 80 100 120 0 10 20 30 40 50 60 70 80 90 100 y x TSP30 (Performance = 941)
- 35. Solution j(Distance = 800) 44 62 69 67 78 64 62 54 42 50 40 40 38 21 35 67 60 60 40 42 50 99 0 20 40 60 80 100 120 0 10 20 30 40 50 60 70 80 90 100 y x TSP30 (Performance = 800)
- 36. Solution k(Distance = 652) 0 20 40 60 80 100 120 0 10 20 30 40 50 60 70 80 90 100 y x TSP30 (Performance = 652)
- 37. Best Solution (Distance = 420) 42 38 35 26 21 35 32 7 38 46 44 58 60 69 76 78 71 69 67 62 84 94 0 20 40 60 80 100 120 0 10 20 30 40 50 60 70 80 90 100 y x TSP30 Solution (Performance = 420)
- 38. 38
- 39. 39 Computer Vision: The world is composed of three-dimensional objects, but the inputs to the human eye and computers' TV cameras are two dimensional. Some useful programs can work in two dimensions, but full computer vision requires partial three-dimensional information that is not just a set of two-dimensional views. At present there are only limited ways of representing three-dimensional information directly, and they are not as good as what humans evidently use.
- 40. 40
- 41. 41 The Poseidon system is based on a network of overhead and underwater cameras installed in a public pool. all linked to a computer system that is going to acquire video signals in real-time filters them extracts human body shapes from images and assesses the movement of these bodies. Whenever the system detects that a body movement pattern (or lack thereof) resembles one of a drowning swimmer, it sends an alert to lifeguards through pagers that indicate the location of the endangered person. Computer Vision applied System
- 42. 42
- 43. 43 • In the 1990s, computer speech recognition reached a practical level for limited purposes. Thus United Airlines has replaced its keyboard tree for flight information by a system using speech recognition of flight numbers and city names. It is quite convenient. Speech recognition application • Telephone-based Information (directions, air travel, banking, etc) • Hands-free (in car) • Second language ('L2') (accent reduction) • Audio archive searching Speech recognition
- 44. 44
- 45. 45 Complex example used speech recognition
- 46. • The process of building expert systems is often called knowledge engineering. The knowledge engineer is involved with all components of an expert system: 46 Expert Systems Building expert systems is generally an iterative process. The components and their interaction will be refined over the course of numerous meetings of the knowledge engineer with the experts and users. We shall look in turn at the various components.
- 47. 47
- 48. 48 Goal: To create computational models of language in enough detail that you could write computer programs to perform various tasks involving natural language. Scientific: to explore the nature of linguistic communication Practical: to enable effective human-machine communication Just getting a sequence of words into a computer is not enough. Parsing sentences is not enough either. The computer has to be provided with an understanding of the domain the text is about, and this is presently possible only for very limited domains. Understanding Natural Language
- 49. 49
- 50. 50
- 51. 51
- 52. 52 AI Branches Representation Knowledge needs to be represented somehow – perhaps as a series of if-then rules, as a frame based system, as a semantic network, or in the connection weights of an artificial neural network. Learning Automatically building up knowledge from the environment – such as acquiring the rules for a rule based expert system, or determining the appropriate connection weights in an artificial neural network. (Detailed in next chapters) Rules These could be explicitly built into an expert system by a knowledge engineer, or implicit in the connection weights learnt by a neural network. Search This can take many forms – perhaps searching for a sequence of states that leads quickly to a problem solution, or searching for a good set of connection weights for a neural network by minimizing a fitness function.
- 53. 53
- 54. 54
- 55. 55 Game playing Game playing is a search problem Defined by: – Initial state – Successor function – Goal test – Path cost / utility / payoff function Characteristics of game playing: • Initial state: initial board position and player • Operators: one for each legal move • Terminal states: a set of states that mark the end of the game • Utility function: assigns numeric value to each terminal state • Game tree: represents all possible game scenarios
- 56. 56 (Our) Basis of Game Playing: Search for best move every time Initial Board State Board State 2 Board State 3 Board State 4 Board State 5 Search for Opponent Move 1 Moves 2 Search for Opponent Move 3 Moves
- 57. 57 May, 1997: Deep Blue beats the World Chess Champion I could feel human-level intelligence across the room vs. You can buy machines that can play master level chess for a few hundred dollars. There is some IS in them, but they play well against people mainly through brute force computation looking at hundreds of thousands of positions. To beat a world champion by brute force and known reliable heuristics requires being able to look at 200 million positions per second.
- 58. What Can AI Do? From these examples • Play a game of table tennis? • Drive safely along a road with signals? • Drive safely along any road? • Buy a week's worth of groceries on the web? • Buy a week's worth of groceries at Berkeley Bowl? • Discover and prove a new mathematical theorem? • Converse successfully with another person for an hour? • Perform a complex surgical operation? • Unload a dishwasher and put everything away? • Translate spoken English into spoken Arabic in real time? • Write an intentionally funny story? 58