Artifical intelligence first lecture notes.pptx

notes on first session of AI

1. AI and ES Session 1 Slides

2. Module Objectives By the end of this module, you should be able to Understand the concepts and practical applications of Artificial Intelligence (AI), Appreciate the contribution of AI in technological development in expert systems, Understand AI new application areas such as Artificial intelligence 2.0

3. What is AI? Artificial intelligence (AI) , Major AI researchers and textbooks define this field as "the study and design of intelligent agents", where an intelligent agent is a system that perceives its environment and takes actions that maximize its chances of success. John McCarthy, who coined the term in 1955, defines it as "the science and engineering of making intelligent machines. AI research is also divided by several technical issues. Some subfields focus on the solution of specific problems. Others focus on one of several possible approaches or on the use of a particular tool or towards the accomplishment of particular applications. The central problems (or goals) of AI research include reasoning, knowledge, planning, learning, natural language processing (communication), perception and the ability to move and manipulate objects  

4. Components of AI: Learning There are a number of different forms of learning as applied to artificial intelligence. The simplest is learning by trial and error. For example, a simple computer program for solving mate-in-one chess problems might try moves at random until mate is found. The program might then store the solution with the position so that the next time the computer encountered the same position it would recall the solution. . 

5. Reasoning To reason is to draw inferences appropriate to the situation. Inferences are classified as either deductive or inductive. An example of the former is, “Fred must be in either the museum or the café. He is not in the café; therefore he is in the museum,” and of the latter, “Previous accidents of this sort were caused by instrument failure; therefore this accident was caused by instrument failure.” Deductive reasoning is a type of deduction used in science and in life. It is when you take two true statements, or premises, to form a conclusion. For example, A is equal to B. B is also equal to C. Given those two statements, you can conclude A is equal to C using deductive reasoning.

6. Reasoning In inductive inference, we go from the specific to the general. We make many observations, discern a pattern, make a generalization, and infer an explanation or a theory An example of inductive logic is, "The coin I pulled from the bag is a penny. That coin is a penny. A third coin from the bag is a penny. Therefore, all the coins in the bag are pennies.“ Abductive reasoning Another form of scientific reasoning that doesn't fit in with inductive or deductive reasoning is abductive . Abductive reasoning usually starts with an incomplete set of observations and proceeds to the likeliest possible explanation for the group of observations

7. Reasoning Abductive reasoning It is based on making and testing hypotheses using the best information available. It often entails making an educated guess after observing a phenomenon for which there is no clear explanation.  For example, a person walks into their living room and finds torn up papers all over the floor. The person's dog has been alone in the room all day. The person concludes that the dog tore up the papers because it is the most likely scenario. Now, the person's sister may have brought by his niece and she may have torn up the papers, or it may have been done by the landlord, but the dog theory is the more likely conclusion. Abductive reasoning is useful for forming hypotheses to be tested. Abductive reasoning is often used by doctors who make a diagnosis based on test results and by jurors who make decisions based on the evidence presented to them

8. Problem solving Problem solving, particularly in artificial intelligence, may be characterized as a systematic search through a range of possible actions in order to reach some predefined goal or solution. Problem-solving methods divide into special purpose and general purpose. A special-purpose method is tailor-made for a particular problem and often exploits very specific features of the situation in which the problem is embedded. In contrast, a general-purpose method is applicable to a wide variety of problems.

9. Perception In perception the environment is scanned by means of various sensory organs, real or artificial, and the scene is decomposed into separate objects in various spatial relationships. Analysis is complicated by the fact that an object may appear different depending on the angle from which it is viewed, the direction and intensity of illumination in the scene, and how much the object contrasts with the surrounding field. At present, artificial perception is sufficiently well advanced to enable optical sensors to identify individuals, autonomous vehicles to drive at moderate speeds on the open road, and robots to roam through buildings collecting empty soda cans. One of the earliest systems to integrate perception and action was FREDDY, a stationary robot with a moving television eye and a pincer hand, constructed at the University of Edinburgh, UK

10. Perception FREDDY was able to recognize a variety of objects and could be instructed to assemble simple artifacts, such as a toy car, from a random heap of components.

11. Language Artificial intelligence can grasp the meaning of simple language, and speak back to you, but it is limited by its literal interpretations of our questions. A computer can know the definition of a word, but it doesn’t understand the meaning of words within a larger context. If you’re interested in tech or sci-fi, you’ve probably heard of the Turing test. Alan Turing was one of the first people to take the potential of AI seriously, and he knew that one day machines would match human intelligence. He had an idea for a simple test: If a human can’t distinguish between a machine and another human in conversation, then the machine has reached the level of human intelligence.

12. Language Chatbots have come a long way from often useless dummies to intelligent assistants that can trick you into thinking you’re actually communicating with a real person. With chatbots getting smarter, people have started using them in learning foreign languages. All you have to do is engage in a dialog with an AI bot and learn through the process of communication. AI-powered language learning chatbots provide customized answers in response to your messages and can even grade your performance or give tips on what you need to improve.  And the best part? You don’t have to face the anxiety of failure that you might when you’re talking to a real person.

13. Alan Turing And The Beginning Of AI Theoretical work The earliest substantial work in the field of artificial intelligence was done in the mid-20th century by the British logician and computer pioneer Alan Mathison Turing Turing often discussed how computers could learn from experience as well as solve new problems through the use of guiding principles—a process now known as heuristic problem solving. Turing gave quite possibly the earliest public lecture (London, 1947) to mention computer intelligence, saying, “What we want is a machine that can learn from experience,” and that the “possibility of letting the machine alter its own instructions provides the mechanism for this.” In 1948 he introduced many of the central concepts of AI in a report entitled “Intelligent Machinery.” However, Turing did not publish this paper, and many of his ideas were later reinvented by others.

14. AI domains

15. AI domains Expert Systems is an interactive and reliable computer-based decision-making system which uses both facts and heuristics to solve complex decision-making problems. It is considered at the highest level of human intelligence and expertise. The purpose of an expert system is to solve the most complex issues in a specific domain. The Expert System can resolve many issues which generally would require a human expert. It is based on knowledge acquired from an expert. It is also capable of expressing and reasoning about some domain of knowledge Characteristics of Expert Systems High performance Understandable Reliable Highly responsive Examples − Flight-tracking systems, Clinical systems.

16. AI domains Expert Systems

17. AI domains Natural Language Processing When you take AI and focus it on human linguistics, you get NLP. Natural language processing makes it possible for computers to extract keywords and phrases, understand the intent of language, translate that to another language, or generate a response. While NLP may get the most attention in consumer applications today, it has significant implications for organizational IT. “Understanding language and communication in general is huge for the enterprise as we spend most of our day communicating in one form or another

18. AI domains Natural Language Processing Any area of the business where natural language is involved may be fodder for the deployment of NLP capabilities. Think chatbots , social media feeds, emails, or complex documentation like contracts or claims forms. NLP is typically deployed to categorize content, extract content, analyze sentiment, summarize documents, translate languages, deploy voice-driven or chat-driven interfaces However, NLP applications come with the same risks of failure as any other AI deployments: Most notably, they can suffer from inflated expectations, unclear business cases, and lack of training data. Additionally, NLP opportunities may require entirely different training sets depending on the language being processed and the context

19. AI domains Natural Language Processing Also, there are many problems in trying to make a computer understand people. Four problems arise that can cause misunderstanding: (1) Ambiguity—confusion over what is meant due to multiple meanings of words and phrases. (2) Imprecision—thoughts are sometimes expressed in vague and inexact terms. (3) Incompleteness—the entire idea is not presented, and the listener is expected to "read between the lines." (4) Inaccuracy—spelling, punctuation, and grammar problems can obscure meaning.

20. AI domains Robotics While AI can be entirely software, robots are physical machines that move. Robots are subject to physical impact, typically through “sensors”, and they exert physical force onto the world, typically through “actuators”, like a gripper or a turning wheel. Accordingly, autonomous cars or planes are robots, and only a minuscule portion of robots is “humanoid” (human-shaped), like in the movies.

21. AI domains Robotics Artificially intelligent robots are the bridge between robotics and AI. These are robots that are controlled by AI programs. Most robots are not artificially intelligent. Up until quite recently, all industrial robots could only be programmed to carry out a repetitive series of movements which, as we have discussed, do not require artificial intelligence. However, non-intelligent robots are quite limited in their functionality.

22. AI domains Robotics A warehouse robot might use a path-finding algorithm to navigate around the warehouse. A drone might use autonomous navigation to return home when it is about to run out of battery. A self-driving car might use a combination of AI algorithms to detect and avoid potential hazards on the road. These are all examples of artificially intelligent robots.

23. AI domains Artificial Neural Networks (ANNs) An artificial neural network (ANN) is the piece of a computing system designed to simulate the way the human brain analyzes and processes information. It is the foundation of artificial intelligence (AI) and solves problems that would prove impossible or difficult by human or statistical standards. ANNs have self-learning capabilities that enable them to produce better results as more data becomes available An artificial neural network (ANN) is the component of artificial intelligence that is meant to simulate the functioning of a human brain. Processing units make up ANNs, which in turn consist of inputs and outputs. The inputs are what the ANN learns from to produce the desired output.

24. AI domains Artificial Neural Networks (ANNs) An ANN initially goes through a training phase where it learns to recognize patterns in data, whether visually, aurally, or textually. During this supervised phase, the network compares its actual output produced with what it was meant to produce—the desired output. The difference between both outcomes is adjusted using back propagation. This means that the network works backward, going from the output unit to the input units to adjust the weight of its connections between the units until the difference between the actual and desired outcome produces the lowest possible error.

25. AI domains Artificial Neural Networks (ANNs) Artificial neural networks have been applied in all areas of operations. Email service providers use ANNs to detect and delete spam from a user’s inbox; asset managers use it to forecast the direction of a company’s stock;  Credit rating firms use it to improve their credit scoring methods; e-commerce platforms use it to personalize recommendations to their audience; chatbots are developed with ANNs for natural language processing; deep learning algorithms use ANN to predict the likelihood of an event;.

26. AI domains Fuzzy Logic Systems (FLS) produce acceptable but definite output in response to incomplete, ambiguous, distorted, or inaccurate (fuzzy) input Fuzzy Logic (FL) is a method of reasoning that resembles human reasoning. The approach of FL imitates the way of decision making in humans that involves all intermediate possibilities between digital values YES and NO. The inventor of fuzzy logic, observed that unlike computers, the human decision making includes a range of possibilities between YES and NO, such as

27. AI domains

28. AI domains

29. Task domains of AI

30. Task domains of AI Humans learn  mundane (ordinary) tasks  since their birth. They learn by perception, speaking, using language, and locomotives. They learn Formal Tasks and Expert Tasks later, in that order. For humans, the mundane tasks are easiest to learn. The same was considered true before trying to implement mundane tasks in machines. Earlier, all work of AI was concentrated in the mundane task domain. Later, it turned out that the machine requires more knowledge, complex knowledge representation, and complicated algorithms for handling mundane tasks. This is the reason  why AI work is more prospering in the Expert Tasks domain  now, as the expert task domain needs expert knowledge without common sense, which can be easier to represent and handle.

31. Intelligent systems Intelligence of a system is the ability of a system to calculate, reason, perceive relationships and analogies, learn from experience, store and retrieve information from memory, solve problems, comprehend complex ideas, use natural language fluently, classify, generalize, and adapt new situations

32. AI Perspectives Artificial intelligence can be viewed from a variety of perspectives. From the perspective of intelligence artificial intelligence is making machines "intelligent" -- acting as we would expect people to act. The inability to distinguish computer responses from human responses is called the Turing test. A  Turing Test  is a method of inquiry in artificial intelligence (AI) for determining whether or not a computer is capable of thinking like a human being. The  test  is named after Alan  Turing , the founder of the Turning  Test Intelligence requires knowledge Expert problem solving - restricting domain to allow including significant relevant knowledge

33. Business Perspectives From a business perspective AI is a set of very powerful tools, and methodologies for using those tools to solve business problems eg customer care relationship management Enterprise seems to be entering a new era ruled by data. What was once the realm of science fiction, AI in business intelligence is evolving into everyday business as we know it. Companies can now use machines algorithms to identify trends and insights in vast reams of data and make faster decisions that potentially position them to be competitive in real-time

34. Programming perspective From a programming perspective, AI includes the study of symbolic programming, problem solving, and search. Typically AI programs focus on symbols rather than numeric processing. Problem solving - achieve goals. Search - seldom access a solution directly. Search may include a variety of techniques. AI programming languages include: – LISP, developed in the 1950s,

35. Techniques used in AI Even apparently radically different AI systems (such as rule based expert systems and neural networks) have many common techniques. Four important ones are: Knowledge Representation: Learning Rules Search

36. Techniques in detail Knowledge 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. A weight represent the strength of the connection between units. If the weight from node 1 to node 2 has greater magnitude, it means that neuron 1 has greater influence over neuron 2 A weight decides how much influence the input will have on the output 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.

37. Techniques in detail: Recap Knowledge Representation: if-then rules , if x is A then y is B x is A ” is called the premise premise (proposition from which another is inferred) “ y is B ” is called the conclusion If the road is slippery, then driving is dangerous. ‹ If an apple is red, then it is ripe. ‹ If the speed is high, then apply the brake a little

38. Techniques in detail: Recap Knowledge Representation: as a frame based system A frame provides a structured representation of an object or a class of objects Example: How to represent: “Car #15 is green.” Solution 1: Green(car15). There are many different ways of representing the same knowledge. what is the color of car15? Solution 2: Color (car15, green).

39. Techniques in detail: Recap Knowledge Representation: as a frame based system What property of car15 has value green?” Solution 3: Prop(car15, color , green).

40. Techniques in detail: Recap Knowledge Representation: as a semantic network A semantic network approach views the meaning of concepts as being determined by their relations to other concepts. Concepts are represented as nodes with labeled links (e.g., IS-A or Part-of) as relationships among the nodes. Thus, knowledge is a combination of information about concepts and how those concepts relate to each other.

41. Techniques in detail: Recap Knowledge Representation: as a semantic network

42. Techniques in detail: Recap Knowledge Representation: as connection weights of an artificial neural network Weight is the parameter within a neural network that transforms input data within the network's hidden layers. A neural network is a series of nodes. Within each node is a set of inputs, weight, and a bias value. As an input enters the node, it gets multiplied by a weight value and the resulting output is either observed, or passed to the next layer in the neural network. Often the weights of a neural network are contained within the hidden layers of the network .

43. Techniques in detail: Recap Knowledge Representation: as connection weights of an artificial neural network   Weights can be thought of as the strength of the connection. Weight affects the amount of influence a change in the input will have upon the output. A low weight value will have no change on the input, and alternatively a larger weight value will more significantly change the output .

44. Techniques in detail Rule Systems : 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

45. AI and related fields Logical AI- What a program knows about the world in general the facts of the specific situation in which it must act, and its goals are all represented by sentences of some mathematical logical language. Search- AI programs often examine large numbers of possibilities, e.g. moves in a chess game or inferences by a theorem proving program Pattern Recognition- When a program makes observations of some kind, it is often programmed to compare what it sees with a pattern. For example, a vision program may try to match a pattern of eyes and a nose in a scene in order to find a face

46. AI and related fields Representation- Facts about the world have to be represented in some way. Usually languages of mathematical logic are used. Inference - From some facts, others can be inferred. Mathematical logical deduction is adequate for some purposes, but new methods of non-monotonic inference have been added to logic since the 1970s. The simplest kind of non-monotonic reasoning is default reasoning in which a conclusion is to be inferred by default, but the conclusion can be withdrawn if there is evidence to the contrary. For example, when we hear of a bird, we man infer that it can fly, but this conclusion can be reversed when we hear that it is a penguin Ordinary logical reasoning is monotonic in that the set of conclusions reached cannot be invalidated even if new facts were added.

47. AI and related fields Common sense knowledge and reasoning This is the area in which AI is farthest from human-level, in spite of the fact that it has been an active research area since the 1950s. While there has been considerable progress, e.g. in developing systems of non-monotonic reasoning and theories of action, yet more new ideas are needed. Learning from experience- Programs do that. The approaches to AI based on connectionism and neural nets specialize in that. connectionist models usually take the form of neural networks, which are composed of a large number of very simple components wired together. Neural network models were inspired by and resemble the anatomy and physiology of the nervous system. Key aspects of many neural network models are that they are able to learn and their behavior improves with training or experience.   

48. AI and related fields Planning- Planning programs start with general facts about the world (especially facts about the effects of actions), facts about the particular situation and a statement of a goal. From these, they generate a strategy for achieving the goal. In the most common cases, the strategy is just a sequence of actions

49. Search and Control Strategies Problem solving is an important aspect of Artificial Intelligence. A problem can be considered to consist of a goal and a set of actions that can be taken to lead to the goal. At any given time, we consider the state of the search space to represent where we have reached as a result of the actions we have applied so far Search is a method that can be used by computers to examine a problem space like this in order to find a goal. Often, we want to find the goal as quickly as possible or without using too many resources A problem space can also be considered to be a search space because in order to solve the problem, we will search the space for a goal state

50. The Importance of Search in AI It has already become clear that many of the tasks underlying AI can be phrased in terms of a search for the solution to the problem at hand. Many goal based agents are essentially problem solving agents which must decide what to do by searching for a sequence of actions that lead to their solutions.

51. Search algorithms important Factors Search algorithm depend on the problem domain? There are four important factors to consider: Completeness – Is a solution guaranteed to be found if at least one solution exists? Optimality – Is the solution found guaranteed to be the best (or lowest cost) solution if there exists more than one solution? Time Complexity – The upper bound on the time required to find a solution, as a function of the complexity of the problem. Space Complexity – The upper bound on the storage space (memory) required at any point during the search, as a function of the complexity of the problem

52. State Space Representations Intelligence is all about making good choices. Making good choices is about considering alternatives, and the effects of the alternatives. The process of systematically considering alternatives is called "search" Hypothesis: All problems that require intelligence can be characterized as a state space and intelligence can be characterized a search in that space. State space- Definition of a problem: State- a condition or mode of the problem Initial state- the start state from which the program tries to solve the problem. Set of operators- an operator is an action that can be taken within the framework of the problem that changes the current state to some other  valid  state in the problem. Goal state- the state we want the problem to be in.

53. State Space Representations State space search is a process in which successive configurations or states of an instance are considered, with the intention of finding a goal state with a desired property. Before an AI problem can be solved it must be represented as a state space. The state space is then searched to find a solution to the problem. A state space essentially consists of a set of nodes representing each state of the problem, arcs between nodes representing the legal moves from one state to another, an initial state and a goal state. Each state space takes the form of a tree or a graph.

54. Classic AI Problems: Homework Travelling Salesman Problem (TSP):   Given a set of cities and distance between every pair of cities, the problem is to find the shortest possible route that visits every city exactly once and returns to the starting point. Towers of Hanoi Tower of Hanoi is mathematical game puzzle where we have   three pile (pillars) and n numbers of disk. This game has some rules  (Rules of game) Only one disk will move at a time. The larger disk should always be on the bottom and the smaller disk on top of it. Move only the uppermost disk. All disk move to destination pile from source pile Task Choose one problem and how AI algorithms may be be applied to solve the classic problem

55. Homework Depth First Search Depth Limited Search Depth First Iterative Deepening Search

56. Future of AI Despite recent advances in Artificial Intelligence (AI) that enable it to win games and drive cars, there are countless untapped opportunities for advanced technology to have a significant and beneficial impact on the world. Particularly so at the intersection of AI and robotics Data, measured in billions of gigabytes every day, is collected by networked devices in virtually every industry. As a result, AI is being tightly interwoven into almost every aspect of our lives, from our cars and medical devices to robots and entertainment. It’s here to stay. The question is, how can we shape it?

57. Future of AI Cognitive AI. Cognitive AI agents are insightful advisors with powerful reasoning engines that can be consulted by people to solve complex problems. Like humans, they are able to fill in these knowledge gaps – understanding a situation based on either prior knowledge or by amalgamating fragmented evidence into a mental model.  Large fleets of ships currently operate largely unmonitored and un-instrumented, especially compared to other modes of transportation such as jets and smart cars

58. Future of AI Cognitive AI.

59. Future of AI

60. Future of AI Pervasive Knowledge refers to a massive  knowledge  generated from multi sources of information such as social network, smart mobile devices, and structure  knowledge  management database. Autonomous everything refers to a system or function is a closed loop (“sense-think-act”). The machine receives information from its environment through sensors (“sense”) processes these data with control software (“think”) based on its analysis, performs an action (“act”) without further human intervention. Other systems initiate or adjust their actions or performance based on feedback from the environment (“automated”) and more sophisticated systems combine environmental feedback with the system’s own analysis regarding its current situation (“autonomous”). Increasing autonomy is generally equated with greater adaptation to the environment and is sometimes presented as increased “intelligence” – or even “artificial intelligence” – for a particular task.

61. Future of AI AI enhanced organization: Enterprise cognitive computing (ECC) — the use of AI to enhance business operations — involves embedding algorithms into applications that support organizational processes. ECC applications can automate repetitive, formulaic tasks and, in doing so, deliver orders-of-magnitude improvements in the speed of information analysis and in the reliability and accuracy of outputs. For example, ECC call center applications can answer customer calls within 5 seconds on a 24-7-365 basis, accurately address their issues on the first call 90% of the time, and transfer complex issues to employees, with less than half of the customers knowing that they are interacting with a machine

62. Future of AI Enhancing the human experience: focusses on human-centered artificial intelligence, that strives to mimic not just a brain but a mind; not just problem-solving but having initiative. By engaging this duality, we amplify human experience with artificial intelligence. Whether motivated by communication, entertainment, or learning, experiences are driven by perception, emotion, and thought. We’re on a mission to push the boundaries of AI interfaces, developing approaches that can solve complex problems by interacting with and observing people. AI will be most impactful when it is accessible to all and built with empathy and ethics in mind. People are naturally social; machines are not

63. Future of AI Enhancing the human experience: The Human Centered AI is built on three foundational pillars: User  -- to be understood through behavior, emotions and preferences. Aya  -- an intelligent agent characterized in similar terms to the user, making it more accessible. These include personality, memory, knowledge, and the ability to interact with humans and machines. Environment  -- which dictates context and provides boundaries of interaction. The interplay between these three pillars establishes a feedback loop between the user and the experience.

64. Industry 4.0   Although the terms "industry 4.0" and "fourth industrial revolution" are often used interchangeably, "industry 4.0" factories have machines which are augmented with wireless connectivity and sensors, connected to a system that can visualise the entire production line and make decisions on its own Cyber-physical systems IoT Cloud computing Cognitive computing

65. Industry 4.0   By Cyber Physical Systems (CPS), we refer to computer–human networks, controlling physical processes, where physical processes affect computations and vice versa. One modern version of Cyber Physical Systems is the Internet of Things ( IoT ). This means they are capable of autonomously functioning based on their physical surroundings he Internet of Things ( IoT ) is one step forward in the advancement of AI in machines and represents a system of interrelated computing devices, capable of operating without human-to-human or human-to-computer interaction. The Industrial Internet of Things ( IIoT ) in this study refers to sensors and other devices networked with industrial applications, enabling data collection, exchange, and analysis, with the objective for increase in productivity, efficiency and economic benefits

66. Homework   Examine the Role of Artificial Intelligence in Cloud Computing

67. End of session 1 Any Questions