ARTIFICIAL INTELLIGENCE VIVA QUESTIONS 1.

Definitions • Agent • Constraint Satisfaction • Expert system • Game playing • Genetic Algorithm and genetic programming • Knowledge acquisition • Knowledge representation • Logic programming prolog • Machine learning • Natural language processing • Robotics • Simulated annealing • Uncertainty • Vision • Search and game playing • Logic • Planning • Probabilistic reasoning • Nonmonotonic Reasoning and truth maintenance systems (TMS) 2. A*: A search algorithm to find the shortest path through a search space to a goal state using a heuristic. See ‘search’, ‘problem space’, admissibility’ and ‘heuristic’. 3. Admissibility: An admissible search algorithm is one that is guaranteed to find the optimal path fron the start node to a goal node, if one exists. In A* search, an admissible heuristic is one that never overestimated the distance remaining from the current node to the goal. 4. Agent “Anything that can be viewed a perceiving its environment through sensors and acting upon that environment through effectors.” [Russel, Norvig 1995] 5. AI: A three-toed sloth of genus Bradypus. This forest-dwelling animal eats the leaves of the trumpet- tree and sounds a high-pitched squeal when disturbed.(Based on the Random House Dictionary definition.) 6. Alpha-Beta Prunning: A method of limiting search in the MiniMax algorithm. The coolest thing you learn in an undergraduate course. 7. Backward Chaining:

See forward chaining 8. Open World Assumption: If there is no idea about a query it is assumed to be unknown. Used by Minimax when the game tree is too large to be searched exhaustively. previously unknown. Belief Network (also Bayesian Network): It is a mechanism for representing probabilistic knowledge. reasoning from a query to the data. The definitions of the Boolean operators are extended to fit this continuous domain. Fuzzy Logic: In Fuzzy logic. Data Mining: Also known as Knowledge Discovery in Databases(KDD) was been defined as “The nontrivial extraction of implicit. 18.. It uses machine learning. it is false. Case. Forward Chaining: In logic system. Computational Linguistics: It is the branch of AI that deals with understanding human language. statistical and visualization techniques to discover and present knowledge in a form which is easily comprehensible to humans. 16. Evaluation Function: A function applied to a game state to generate a guess as to who is winning. See Backward Chaining. 17. 11.. 13. By avoiding discrete truth values.In a logic system. Depth-first Search: It is an uninformed search algorithm where the deepest non-terminal node is expanded first.1]. 12. 10. It is also called natural language processing. 14. 15. Fuzzy logic avoids some of the problems inherent in either-or judgments and . reasoning from facts to conclusions is called forward chaining. truth values are real values in the closed interval [0. Inference algorithms in belief networks use the structure of the network to generate inference efficiently (compared to joint probability distributions over all the variables). and potentially useful information from data” in Frwaley and Piatetsky-Shapiro’s overview. Closed World Assumption: The assumption that if a system has no knowledge about a query.Based Reasoning: Technique whereby “cases” similar to the current problem are retrieved and their “solutions” modified to work on the current problem. Breadth-first Search: It is an uninformed search algorithm where the shallowest node in the search tree is expanded first. 9.

Language Acquisition: A relatively new sub-branch of AI. The ideas of iterative deepening applied to A*. Fuzzy Logic has applications in control theory. Machine Learning: A field of AI concerned with programs that learn. 22. Language acquisitions are a technique for the computer to generate the grammar rules itself. It includes reinforcement Learning and Neural Networks among many other fields. tradionally computational linguists tried to make computers understand human language by giving the computer grammar rule. Natural language (NL): This is used to evolve languages that human use to communicate with one another. Heuristic: The dictionary defines it as a method that serves as an aid to problem solving. 23. 26. 29. Search 20. and leaves unordered any steps that can be done in any order. Nonlinear Planning: A planning paradigm which does not enforce a total (linear) ordering on the components of a plan is called nonlinear planning. Iterative Deepening: It is an uninformed search that combines good properties of depth-first and breadth-first search. 21. Modus Ponens: An inference rules that says: if you know x and you know that ‘If x is true then y is true then you can conclude y’. Natural language Queries: Using human language to get information from a database is called Natural language queries. It si sometimes defined as any ‘rule of thumb’. MiniMax: It is an algorithm for game playing in games with perfect information. 19. Technically. Information Extraction: Getting computer-understandable information from human-readable (ie natural language) documents is called Information Extraction. 25. Partial order Planner. See also: Admissibility. . 27.yields natural interpretations of utterances like “very hot”. a heuristic is a function that takes a state as input and output a value for that state often as a guess of how far away that state is from the goal state. 28. 24. A planner that only orders steps that need to be ordered.

Unification: The process of finding a substitution (an assignment of constants and variables to variables) that makes two logical statements look the same. A machine that passes the test should certainly be considered intelligent. a general- . More precisely. may be characterized as a systematic search through a range of possible actions in order to reach some predefined goal or solution. 33. See ‘Admissibility’. 34. Strong AI: Claim that computer can be made to actually think. Search: The finding of a path from a start to a goal state. 36. Planning: A field of AI concerned with systems that construct sequences of actions to achieve goal in real-world-like environments. Problem-solving methods divide into special purpose and general purpose.30. such that any implementation of such a program is really thinking. Turing test: A. This test would satisfy most people but not all philosophers. Alan Turing’s 1950 article Computing Machinery and Intelligence discussed conditions for considering a machine to be intelligent. just like human beings do. Weak AI: Claim that computer is important tools in the modeling and simulation of human activity. Validation: The process of confirming that one’s model uses measurable inputs and produces output that can be used to make decisions about the real world. but a machine could still be considered intelligent without knowing enough about humans to imitate a human. and the human would try to persuade the observer that it’s human and the machine would try to fool the observer. Verification: The process of confirming that an implemented model works as intended. 32. 37. The observer could interact with the machine and a human by teletype (to avoid requiring that the machine imitate the appearance or voice of the person). 35. particularly in artificial intelligence. The Turing test is a one-sided test. Problem solving: Problem solving. Problem Space (also state space): It is the formulation of an AI problem into states and operators. A special-purpose method is tailor-made for a particular problem and often exploits very specific features of the situations in which the problem is embedded. There is usually a start state and a goal state. In contrast. ‘Problem space’ and ‘Heuristic’. the claim that there exists a class of computer program. He argues that if the machine could successfully pretend to be human to a knowledgeable observer then you certainly should consider it intelligent.

MOVELEFT and MOVERIGHT – until the goal is reached. is known as rote learning. Reasoning: To reason is to draw inferences appropriate to the situation in hand. form a mini-language. Learning that involves generalization leaves the learner able to perform better in situations not previously encountered. An example of the former is “Fred is either in the museum or the cafZ.g. and so forth. the truth of the premises guarantees the truth of the conclusion. so probably this one was caused by instrument failure”.purpose technique used in AI is means-end analysis. Role learning is relatively easy to implement on a computer. and of the latter “Previous accidents just like this one have been caused by instrument failure. and processes internal to the perceiver analyze the scene into objects and their features and relationships. MOVEBACK. “jump” until presented at least once with “jumped”. 39. 40. especially deductive .a step-by-by. The simple memorizing of individual items. exemplified in statements like ‘Those clouds mean rain’ and the fall in pressure means the valve in malfunctioning’. The difference between the two is that in the deductive case. despite the truth of the premises. For example. reduction of the difference between the current state and the final goal. MOVEFORWARD. so he’s the museum’. More challenging is the problem of implementing what is called generalization. the truth of the premises lends support to the conclusion that the accident was caused by instrument failure. Traffic signs. There has been considerable success in programming computers to draw inferences. PUTDOWN. or incremental.solutions to problems. depending on the angle from which it is viewed. 38. but nevertheless further investigation might reveal that. the hazard-ahead sign means hazard ahead. etc. The program remembers the successful move and next time the computer is given the same problem it is able to produce the answer immediately. whereas a program that is able to generalize from examples can learn the “added” rule. words of vocabulary. 41. a simple program for solving mate-in-one chess problems might try out moves at random until one is found that achieves mate. Learning: Learning is distinguished into a number of different forms. Perception: In perception the environment is scanned by means of various sense organs. and so form the past tense of “jump” in the absence of any previous to generalize complex rules from data. for example. The program selects actions from a list of means in the case of a simple robot this might consists of PICKUP. whether or not parts of it are projecting shadows. A program that learns past tenses of regular English verbs by rote will not be able to produce the past tense of e. Inferences are classified as either deductive or inductive. The simplest is learning b trialand-error. This meaning-by-convention that is distinctive of language is very different from what is called natural meaning. Language-understanding: A language is a system of signs having meaning by convention. it being a matter of convention that. whereas in the inductive cases. Analysis is complicated by the fact that one and the same object may present many different appearance on different occasions. real or artificial. for example. the conclusion is in fact false. he isn’t in the cafZ.

However.Other Search Techniques) 43. Heuristic Search (Heuristics. 44.inferences. . 42.Heuristic Search. means-end analysis: It is an AI Technique which tries to reduce the “difference” between a current state and a goal state.Best-First Search – hill climbing – Minimizing Cost – A* Search [A-Start Search]. Generate and test: It is a search technique which proposes possible solutions and then tests them for their feasibility. 45. Heuristic Search: It is an informed method of searching a state space with the purpose of reducing its size and finding one or more suitable goal states. a program cannot be said to reason simply in virtue of being able to draw inferences. Reasoning involves drawing inferences that are relevant to the task or situation in hand. One of the hardest problems confronting AI is that of giving computers the ability to distinguish the relevant from the irrelevant.

Does AI aim to put the human mind into the computer? • Some researchers say they have that objective. • The human mind has a lot of peculiarities. What is intelligence? • Intelligence is the computational part of the ability to achieve goals in the world. • Varying kinds and degrees of intelligence occur in people. • By the late 1950s. When did AI research start? • The English mathematician Alan Turing gave a lecture on it in 1947. 7. What is artificial intelligence? • It is the science and engineering of making intelligent machines. • On the one hand. 6. and I’m not sure anyone is serious about imitating all of them. What about IQ? • IQ is based on the rates at which intelligence develops.ARTIFICIAL INTELLIGENCE AND EXPERT SYSTEMS VIVA QUESTIONS 1. and most of them were basing their work on programming computers. . there were many researches on AI. 5. 8. especially intelligent computer programs. • AI doesn’t have to confine itself to methods that are biologically observable. but a machine could still be considered intelligent without knowing enough about humans to imitate a human. Does AI aims at human-level intelligence? • Yes. The ultimate effort is to make computer programs that can solve problems and achieve goals in the world as well as humans. 4. many animals and some machines. 3. • It is related to the similar task of using computers to understand human intelligence. • A machine that passes the test should certainly be considered intelligent. What is the Turing Test? • The Turing test is a one-sided test. but may be they are using the phrase metaphorically. 2. He also may have been the first to decide that AI was best researched by programming computers rather than by building machines. we can learn something about how to make machines solve problems by observing other people or just by observing our own methods • AI researchers are free to use methods that are not observed in people or that involve much more computing than people can do. Isn’t AI about simulating human intelligence? • Sometimes but not always or even usually.

11. • But unlike objects. an agent is defined in terms of its behavior . To beat a world champion by brute force and known reliable heuristics requires being able to look at 200 million positions per second. which are defined in terms of methods and attributes. • Heuristic classification. • Expert systems • Computer vision • Understanding natural language • Speech recognition 10. and this is presently possible only for very limited domains. What is speech recognition? • In the 1990s. 12. What is an agent? • The term “agent” describes a software subtraction. What are the applications of AI? • Game playing. computer speech recognition reached a practical level for limited purposes. 13. What do you mean by an Expert system? • An expert system. • The computer has to be provided with an understanding of the domain the text is about. They play well against people mainly through brute force computation looking at hundreds of thousands of positions. Explain about understanding natural language applications in AI? • Just getting a sequence of words into a computer is not enough. • The concept of an agent provides a convenient and powerful way to describe a complex software entity that is capable of acting with a certain degree of autonomy in order to accomplish tasks on behalf of its user. also know as a knowledge based system • It is a computer program that contains some of the subject-specific knowledge. Thus united airlines have replaced its keyboard tree for flight information by system using speech recognition of flight numbers and city names. 14. • It is possible to instruct some computers using speech. similar to OOP terms such as methods. Parsing sentences is nto enough either. functions and objects. most users have gone back to the keyboard and the mouse as still more convenient.9. and contains the knowledge and analytical skills of one or more human experts. What is game playing? Game playing • • • You can buy machines that can play master level chess for a few hundred dollars. an idea. or a concept. • The class of program was first developed by researchers in artificial intelligence during the 1960s and 1970s and applied commercially throughout the 1980s.

19. What is DFS? Depth-first search (DFS) is an algorithm for traversing or searching a tree. intelligent or social behaviour. 18. 20. • The A* algorithm is therefore an example of best-first search. What is uncertainty? • The lack of certainty. What is A* algorithm? • A* (pronounced “A star” is a graph/ tree search algorithm that finds a path from a given initial node to a given goal node (or one passing a given goal test). • Cognitive models are used to study e. • It is the use of computers to model cognitive behaviour and sometimes the study of cognitive behavior to improve the usage of computers. 23. and emergent properties of connectionist architecture.g. A state of having limited knowledge where it is impossible to exactly describe existing state or future outcome. Thus the term ‘Neural Network’ has two • Biological Neural Networks. What is Artificial Neural Networks? • Artificial Neural Networks are made up of interconnecting artificial neurons. 16. The modern usage of the term often refers to artificial neurons or nodes. it explores their unexplored neighbour nodes. • Artificial Neural Networks may either be used to gain an understanding of biological neural networks. • Then for each of those nearest nodes. more than one possible outcome. 17. • It employs a “heuristic estimate” h(x) that ranks each node x by an estimate of the best route that goes through that node. What is Cognitive Modeling? • A cognitive model is a model of cognitive processes. What do you mean by BFS? • Breadth-first search (BFS) is a graph search algorithm that begins at the root node and explores all the neighboring nodes. tree structure or graph. What do you mean by a Heuristic? • It can be either be any algorithm that gives up finding the optimal solution for an improvement in run time • It can be a function that estimates the cost of the cheapest path from one node to another. • Artificial Neural Networks. 21.15. What do you mean by Neural Networks? Neural network had been used to refer to a network or circuitry of biological neurons. and so on. until it finds the goal .

Sign up to vote on this title
UsefulNot useful