Questions and Answers of Artificial Intelligence Structures

Implement ID3 in a language of your choice and run it on the credit history example from the text. If you use LISP, consider the algorithms and data structures developed in Section 15.13 for
Develop a simple table of examples in some domain, such as classifying animals by species, and trace the construction of a decision tree by the ID3 algorithm.
Using Shannon’s formula, show whether or not a message about the outcome of a spin of a roulette wheel has more information than one about the outcome of a coin toss. What if the roulette wheel
Using the information theoretic selection function of Section 10.4.3, show in detail how ID3 constructs the tree of Figure 10.14 from examples in Table 10.1. Be sure to show the calculations used in
Build the version space search algorithms in Prolog, or the language of your choice. If you use Prolog or Java, see hints for version space search in the auxiliary materials.
The run of the candidate elimination algorithm shown in Figure 10.9 does not show candidate concepts that were produced but eliminated because they were either overly general, overly specific, or
Create cliques and a junction tree for the following situation (seen in Figure 9.23). Robbery, vandalism and an earthquake can all set off (cause) a house alarm. There is also a measure of the
Create an algorithm for Bayesian belief propagation and apply it to the slippery sidewalk domain of Section 9.3.2. You might use Pearl’s (1988) message passing approach or the clique triangulation
Create another reasoning network similar to that of Figure 9.4 and show the dependency lattice for its premises, as was done in Figure 9.5.Figure 9.4Figure 9.5
Create a new example of diagnostic reasoning and use the Dempster–Shafer equations of Section 9.2.3 to obtain belief distributions as in Table 9.1 and 9.2Table 9.1Table 9.2
Given the following rules in a “back-chaining” expert system application:A ∧ not (B) ⇒ C (.9)C v D ⇒ E (.75)F⇒A (.6)G⇒ D (.8)The system can conclude the following facts (with
Identify three application domains where reasoning under conditions of uncertainty is necessary. Pick one of these areas and design six inference rules reflecting reasoning in that domain.
In Section 8.4.3 we presented a planner created by Nilsson and his students at Stanford (Benson and Nilsson 1995). Teleo-reactive planning allows actions described as durative, i.e., that must
Read the ABSTRIPS research (Sacerdotti 1974) and show how it handles the linearity (or incompatible subgoal) problem in planning.
There were a number of important issues presented near the end of Section 7.4 related to the creation of an agent-oriented solutions to problems. Pick one of these and discuss the issue further.
At the end of Section 7.3.1, there are five potential issues that Brooks’ subsumption architecture (1991a) must address to offer a successful general-purpose approach to problem-solving. Pick one
Each of the following sequences of characters is generated according to some general rule. Describe a representation that could be used to represent the rules or relationships required to continue
Presented the general conflict resolution strategies of refraction, recency, and specificity. Propose and justify two more such strategies.
Consider the financial advisor problem discussed in Chapters 2, 3, and 4. Using predicate calculus as a representation language:a. Write the problem explicitly as a production system.b. Generate
Using the rule in Example 6.2.3 as a model, write the eight move rules needed for the full 8 × 8 version of the knight’s tour.Data from Example 6.2.3We may generalize the knight’s tour solution
Write the pseudo-code definition for a breadth-first version of pattern search. Discuss the time and space efficiency of this algorithm.
Using the move and path definitions for the knight’s tour of Section 6.2.2, trace the execution of pattern search on the goals:a. Path (1,9).b. Path (1,5).c. Path (7,6).When the move predicates
Using the move and path definitions for the knight’s tour of Section 6.2.2, trace the execution of pattern search on the goals:a. Path (1,9).b. Path (1,5).c. Path (7,6).When the move predicates
What is the probability of being dealt the following hands in a five card poker game (from the normal deck of fifty-two cards)?a. A “flush” or all cards from the same suit.b. A “full house”
In a survey of 250 television viewers, 88 like to watch news, 98 like to watch sports, and 94 like to watch comedy. 33 people like to watch news and sports, 31 like to watch sports and comedy, and 35
Trace the goal-driven good-dog problem of Example 3.3.4 in a data-driven fashion.Data from Example 3.3.4This example is taken from the predicate calculus and represents a goal-driven graph search
An alternative semantic model for logic programming is that of Flat Concurrent Prolog. Compare Prolog as seen in Section 14.3 with Flat Concurrent Prolog (Shapiro 1987).Use factoring and resolution
When we go out to dinner, we always like to park as close as possible to the restaurant. Assume the restaurant is situated on a very long street running east to west, which allows parking on one side
Consider the general domain of grid-world navigation tasks, where there is a goal state, obstacles, and a discount factor γ < 1. The actions are stochastic, so the agent may slip into a different
Create a hidden Markov model to predict the half inning score sequence of an American baseball game. Suppose the half inning sequence is 0, 0, 0, 1, 0, 1, 1, 2, 2, 0, 0, 0, 0, 2, 0, 0, 0, 0. For
Discuss the role of techniques such as gray coding for shaping the search space of the genetic algorithm. Describe two other areas where similar techniques may be important.
Prove A* is admissible. The proof should show that:a. A* search will terminate.b. During its execution there is always a node on open that lies on an optimal path to the goal.c. If there is a
Does admissibility imply monotonicity of a heuristic? If not, can you describe when admissibility would imply monotonicity?
A Caesar cipher is an encryption scheme based on cyclic permutations of the alphabet, with the i-th letter of the alphabet replaced by the (i + n)-th letter of the alphabet. For example, in a Caesar
Perform minimax on the tree shown in Figure 4.30.Figure 4.30 DO 3 MAX B A • E F 5 MO N 07 G C KL 5 7 8 4 H
Consider three-dimensional tic-tac-toe. Discuss representational issues; analyze the complexity of the state space. Propose a heuristic for playing this game.
Perform alpha-beta pruning on the tic-tac-toe search of Figures 4.23, 4.24, and 4.25. How many leaf nodes can be eliminated in each case?Figures 4.23Figures 4.24,Figures 4.25 我 6-5=1 5-5-0 6-5=1
a. Create an algorithm for heuristically searching and/or graphs. Note that all descendants of an and node must be solved to solve the parent. Thus, in computing heuristic estimates of costs to a
A fair six-sided die is tossed five times, and the numbers up are recorded in a sequence. How many different sequences are there?
Find the number of distinguishable permutations of the letters in the word MISSISSIPPI, of the letters of the word ASSOCIATIVE.
Suppose that an urn contains 15 balls, of which eight are red and seven are black. In how many ways can five balls be chosen so that:a. All five are red? all five are black?b. Two are red and three
How many ways can a committee of three faculty members and two students be selected from a group of five faculty and seven students?
Two dice are tossed. Find the probability that their sum is:a. 4.b. 7 or an even number.c. 10 or greater.
A card is drawn from the usual fifty-two card deck. What is the probability of:a. Drawing a face card (jack, queen, king or ace).b. Drawing a queen or a spade.c. Drawing a face card or a club.
In an ancient Hindu tea ceremony, there are three participants: an elder, a servant, and a child. The four tasks they perform are feeding the fire, serving cakes, pouring tea, and reading poetry;
Read “What’s in a Link” by Woods (1985). Section IV of this article lists a number of problems in knowledge representation. Suggest a solution to each of these problems using logic, conceptual
Create a model-based reasoning system for a simple electronic device. Combine several small devices to make a larger system. You can use if... then... rules to characterize the system functionality.
Read one of the early papers using model-based reasoning to teach children arithmetic (Brown and Burton 1978) or electronic skills (Brown and VanLehn 1980). Comment on this approach.
In Section 8.2 we introduced a set of rules for diagnosing automobile problems. Identify possible knowledge engineers, domain experts, and potential end users for such an application. Discuss the
Critique the shell you used for Exercise 5. What are its strengths and weaknesses? What would you do to improve it? Was it appropriate to your problem? What problems are best suited to that tool?
a. As presented in the text, best-first search uses the closed list to implement loop detection. What would be the effect of eliminating this test and relying on the depth test, g(n), to detect
Compare the three 8-puzzle heuristics of Figure 4.14 with the heuristic of adding the sum of distances out of place to 2 times the number of direct reversals. Compare them in terms of:Figure
The sliding-tile puzzle consists of three black tiles, three white tiles, and an empty space in the configuration shown in Figure 4.29. The puzzle has two legal moves with associated costs: Figure
Give a heuristic that a block-stacking program might use to solve problems of the form “stack block X on block Y.” Is it admissible? Monotonic?
With the Levenshtein metric of Section 4.1.2, use dynamic programming to determine the minimum edit distance from source strings sensation and excitation to target string execution.
Use the backward component of the dynamic programing algorithm to find another optimal alignment of the characters of Figure 4.6. How many optimal alignments are there?Figure 4.6 BAADD CABD DA 0 1 2
Extend the the “most wins” heuristic for tic-tac-toe two plys deeper in the search space of Figure 4.3. What is the total number of states examined using this heuristic? Would the traditional
A Hamiltonian path is a path that uses every node of the graph exactly once. What conditions are necessary for such a path to exist? Is there such a path in the Königsberg map?
Give the graph representation for the farmer, wolf, goat, and cabbage problem: A farmer with his wolf, goat, and cabbage come to the edge of a river they wish to cross. There is a boat at the
Build a finite state acceptor that recognizes all strings of binary digits:(a) That contain “111.(b) That end in “111.(c) That contain “111” but not more than three consecutive “1s.
Give an instance of the traveling salesperson problem for which the nearest-neighbor strategy fails to find an optimal path. Suggest another heuristic for this problem.
“Hand run” the backtrack algorithm on the graph in Figure 3.29. Begin from state A. Keep track of the successive values of NSL, SL, CS, etc.Figure 3.29 B E FO A C J K L M N D H OPR
Implement a backtrack algorithm in a programming language of your choice.
Determine whether goal-driven or data-driven search would be preferable for solving each of the following problems. Justify your answer.a. Diagnosing mechanical problems in an automobile.b. You
Write a backtrack algorithm for and/or graphs.
Give another example of an and/or graph problem and develop part of the search space.
a. Prove that modus ponens is sound for propositional calculus. Hint: use truth tables to enumerate all possible interpretations.b. Abduction is an inference rule that infers P from P→Q and Q.
A new operator, ⊕, or exclusive-or, may be defined by the following truth table:Create a propositional calculus expression using only ∧, ∨, and ¬ that is equivalent to P ⊕ Q. Prove their
Using truth tables, prove the identities of Section 2.1.2.Data from Section 2.1.2 Q PÂQ T T T T F F F T F F F P FL
The logical operator “↔” is read “if and only if.” P↔Q is defined as being equivalent to (P→Q) ∧ (Q→P). Based on this definition, show that P↔Q is logically equivalent to (P ∨
Prove that implication is transitive in the propositional calculus, that is, that ((P → Q) ∧ (Q → R)) → (P → R).
Attempt to unify the following pairs of expressions. Either show their most general unifiers or explain why they will not unify.a. p (X, Y) and p(a,Z)b. p (X, X) and p(a,b)c. Ancestor (X, Y) and
Give two alternative interpretations for the blocks world description of Figure 2.3.Figure 2.3 C a on(c,a) on(b,d) ontable(a) ontable(d) clear(b) clear(c) hand empty b d
a. Compose the substitution sets {a/X, Y/Z} and {X/W, b/Y}.b. Prove that composition of substitution sets is associative.c. Construct an example to show that composition is not commutative.
Jane Doe has four dependents, a steady income of $30,000, and $15,000 in her savings account. Add the appropriate predicates describing her situation to the general investment advisor of the example
Write a set of logical predicates that will perform simple automobile diagnostics (e.g., if the engine won’t turn over and the lights won’t come on, then the battery is bad). Don’t try to be
The following story is from N. Wirth’s (1976) Algorithms + data structures = programs. I married a widow (let’s call her W) who has a grown-up daughter (call her D). My father (F), who visited us
Create and justify your own definition of artificial intelligence.
Give several other examples of Aristotle’s distinction between matter and form. Can you show how your examples might fit into a theory of abstraction?
Much traditional Western thought has dwelt on the mind–body relationship. Are the mind and body:a. Distinct entities somehow interacting, orb. Is minding an expression of “physical processes”,
Criticize Turing’s criteria for computer software being “intelligent”.
Describe your own criteria for computer software to be considered “intelligent”.
Although computing is a relatively new discipline, philosophers and mathematicians have been thinking about the issues involved in automating problem solving for thousands of years. What is your
Given the differences between the architectures of modern computers and that of the human brain, what relevance does research into the physiological structure and function of biological systems have
Pick one problem area that you feel would justify the energy required to design an expert system solution. Spell the problem out in some detail. Based on your own intuition, which aspects of this
Add two more benefits for expert systems to those already listed in the text. Discuss these in terms of intellectual, social, or financial results.
Discuss why you think the problem of machines “learning” is so difficult.
Discuss whether or not you think it is possible for a computer to understand and use a natural (human) language.
List and discuss two potentially negative effects on society of the development of artificial intelligence technologies.
Derive a resolution proof of the theorem of Figure 14.1.Figure 14.1 R 1. R 2. R 3. R 4. R 5. R 6. R 7. R 8. R 9. R 10. R 11. R 12. A B → BA AVB→BVA AB-B-A A A A AVA A A (BC) (A-B). C Av (BVC) →
Classify each of the following sentences as either syntactically incorrect, syntactically correct but meaningless, meaningful but untrue, or true. Where in the understanding process is each of these
Parse each of these sentences using the “dogs world” grammar of Section 15.2.1. Which of these are illegal sentences? Why?The dog bites the dog.The big dog bites the man.Emma likes the boy.The
Produce a parse tree for each of the following sentences. You will have to extend our simple grammars with more complex linguistic constructs such as adverbs, adjectives, and prepositional phrases.
Extend the dogs world grammar to include adjectives in noun phrases. Be sure to allow an indeterminate number of adjectives. Use a recursive rule, adjective_list, that either is empty or contains
Take the previous problem and put its words, including punctuation, in random order.Data from previous problemExtend the database front end example of Section 15.5.2 so that it will answer questions
As a project, build an information extractions system for some knowledge domain to be used on the WWW. See Section 15.5.3 for suggestions.Data from Section 15.5.3The World Wide Web offers many
From this book’s auxiliary materials, take the structure of the Prolog context free and context sensitive parsers and ad adjectives, adverbs, and propositional phrases to the grammar.
From this book’s auxiliary materials, take the structure of the Prolog probabilistic context free parser and ad adjectives, adverbs, and propositional phrases to the grammar.
What is the probability that a four-digit integer with no beginning zeros:a. Has 3, 5, or 7 as a digit?b. Begins with 3, ends with 5, or has 7 as a digit?
The expectation is the mean or average of the value of a random variable. In throwing a die, for example, it can be calculated by totaling up the resulting values from a large number of throws and

Showing 100 - 200 of 247