Applied Artificial Intelligence An Engineering Approach 2nd Edition Bernhard G Humm - Solutions

Unlock the potential of your learning with our comprehensive solution manual for "Applied Artificial Intelligence: An Engineering Approach, 2nd Edition" by Bernhard G. Humm. Access a wealth of knowledge with step-by-step answers, solved problems, and detailed chapter solutions. Our online resource offers an extensive collection of questions and answers, including an instructor manual and test bank, to support your educational journey. Download solutions in PDF format for free and explore the textbook's full depth. Whether you need an answers key or chapter solutions, our expertly crafted content is designed to enhance your understanding.

Problem 3 Using some of methods from this chapter, reengineer one of the BN applications you developed for one of the problems from Chapter 5. That is, redo your application, but employ techniques from this chapter. Write up what you have done, with reference to specific techniques used. Also,
Problem 2 Consider the BN that you constructed for your own domain in Problem 5, Chapter 2.(Or if you haven’t completed this problem, use an example network from elsewhere in this text, or one that comes with a BN software package.)1. Identify the type of each node: query, observation, context,
Problem 1 The elicitation method of partitioning to deal with local structure is said in the text to directly correspond to the use of classification trees and graphs in Chapter 6. Illustrate this for the Flu, Measles, Age, Fever example by filling in the partitioned CPT with hypothetical numbers.
Run the causal discovery program on some of the data sets at our book web site.Try the program with and without giving it prior information (such as variable order).Evaluate how well it has done using two or more of the measures: predictive accuracy, KL distance, information reward, edit distance,
Obtain or write your own Markov blanket discovery algorithm, after reading some of the references to the literature above. Generate some artificial data from a Bayesian net (e.g., using Netica) and see how well its Markov blankets are discovered from the data.Evaluation Problem For this problem you
Implement the PC algorithm using the c2 from x9.11. Compare the results experimentally with one of the metric causal discovery programs from prior problems.
Implement a simple greedy search through the DAG or TOM space. Test the result on artificial data using one or more metrics from prior problems.
Implement one of the alternative metrics for discrete causal models, namely MDL, or MML (or BDe, after reading Heckerman and Geiger, 1995).
Implement the Bayesian metric PCH(hi;e) of (9.3) for discrete causal models.
Reimplement your oracle-based algorithm, whether IC or PC, using significance testing of partial correlations instead of the oracle, with the input being joint samples across the variables rather than a known Bayesian net. Using a variety of a levels(say, 0.01, 0.05, 0.1), compare the results you
Instead of, or in addition to, the IC algorithm above, implement the PC algorithm, again using the oracle from Chapter 2 (rather than significance tests). Your PC learning algorithm should take as input a list of variables and then, using the oracle supplied with some Bayesian network, discover as
Implement the IC algorithm using your d-separation oracle. That is, your IC learning algorithm should take as input a list of variables and then, using the oracle supplied with some Bayesian network, discover as much of the Bayesian network structure as possible. It should output the network
If you have not already done so, complete the oracle from the problems in Chapter 2. Run your d-separation oracle on the additional test network for the IC learning example, LearningEg.dne, which can be found in http://www.csse.monash.edu.au/bai.
Consider the model:Find all the dags which are Markov equivalent. (You can use the IC algorithm to do this.) b d 00 g h
Consider the model:1. Identify all of the active paths between pairs of variables.2. Use Wright’s decomposition rule (in either form) to generate a system of equations for the correlations in terms of the path coefficients.3. Solve the system of equations — i.e., convert it into a set of
Write an interpretive discussion of your work in Problem 2 (and Problem 3, if you’ve done that). Assess your own evaluation procedure, including your treatment of the variance in your measures.
Extend your work for Problem 2 to include AODE.
Create and parameterize the naive Bayes model for the German credit scoring problem.Compute the following evaluation measures for both the default classifier (i.e., predicting all loans will be repaid, based on the prior probability of a 70% chance of repayment) and your NB model:1. Predictive
Prove that IRB is a proper generalization of IRG — in other words, prove that IRB in the binomial case is equivalent to Good’s information reward (assuming that the prior probability is uniform).
1. Implement maximum likelihood expectation maximization (Algorithm 6.5).2. Implement MAP expectation maximization (Algorithm 6.6).Try your algorithm(s) out using some of the data sets with missing values from the book Web site and report the results. Note that there is Lisp code available on the
Implement the Gibbs Sampling algorithm (6.3) for estimating parameters with incomplete data. The missing Cooper & Herskovits computations can be filled in by copying the Lisp function for them on the bookWeb site, or by skipping forward and implementing Equation (9.3) from Chapter 8. Try your
Implement the multinomial parameterization algorithm (6.1). Test it on some of the data sets on the book Web site and report the results.
In this problem you will analyze a very simple artificially created data set from the book Web site http://www.csse.monash.edu.au/bai.html which was created by a v-structure process X ! Y Z — i.e., one with three variables of which one is the child of the other two which are themselves not
Suppose your prior Dirichlet distribution for the roll of a die is D[2;2;2;2;2;2] and that you update this distribution as in x6.2.2. Roll a die ten times and update this. Is the posterior distribution flat? How might you get a flatter posterior distribution?Experimental Problem
Suppose your prior distribution for the probability of heads of a coin in your pocket is B(2,2). Toss the coin ten times. Assuming you update your distribution as in x6.2.1, what is your posterior distribution? What is the expected value of tossing the coin on your posterior distribution? Select a
Design a Bayesian network bank loan credit system which takes as input details about the customer (banking history, income, years in job, etc.) and details of a proposed loan to the customer (type of loan, interest rate, payment amount per month, etc.)and estimates the probability of a loan default
Build a Bayesian network for diagnosing why a car on the road has broken down.First, identify the most important five or six causes for car breakdown (perhaps, lack of fuel or the battery being flat when attempting to restart). Identify some likely symptoms you can use to distinguish between these
Suppose that a dentist wants to use a Bayesian network to support the diagnosis and treatment of a patient’s toothache. Build a BN model including at least 3 possible causes of toothache, any other symptoms of these causes and two diagnostic tests.Then extend the model to include possible
In x5.6.4, we saw how the use of a sensor status node can be used to model both intermittent and persistent faults in the sensor. In the text this node had only two states, working and defective, which does not represent the different kinds of faults that may occur, in terms of false negatives,
All versions of the Bayesian Poker Player to date have used a succession of four distinct Bayesian networks, one per round. In a DDN model for poker, each round in the poker betting would correspond to a single time slice in the DDN, which would allow explicit representation of the interrelation
Determine a method of estimating the expected future contributions to the pot by both BPP and OPP, to be used in the utilities in the Winnings node. Explain your method together with any assumptions used.
What are the observation nodes when using the Poker network for estimating BPP’s probability of winning? What are the observation nodes when using the network to assess the opponent’s estimate of winning chances?
Build the normative asteroid Bayesian network of Figure 5.19 and put in some likely prior probabilities and CPTs. Make a copy of this network, but increase the prior probability for P4; this will play the role of the user model for this problem. Operate both networks by observing P1, P2 and P3,
Implement the Car Accident OOBN, then extend it (along the lines of Example 4.3 in (Koller and Pfeffer, 1997a)) with subclasses to mode: a fuel-injected engine (a subclass of ENGINE), which is less reliable and is more likely to provide a high power level; a sport-car (a subclass of CAR) with a
Implement the Orchard Farm OOBN. (NB: an unparameterized Hugin version is provided online). Investigate how Mr Bunce’s decision about irrigating the orchard changes when (a) the prior for Drought changes and/or (b) the cost of irrigating changes.
Suppose a student’s final grade for a subject is determined by the difficulty of the subject and the student’s innate intelligence (ignoring the impact of the student’s effort). Build an OOBN to model the situation where each student takes multiple subjects, and each subject has multiple
Implement the Fever DDN (see Figure 4.15) as an OOBN.
Parameterize the Orchard Farming OOBN example given in Section 4.7.3. Investigate the trade-off between the spray and irrigation costs and the return from the crop.See how the decision changes when the probability of drought increase, the disease mutates to become highly virulent, or the tree
Cate brought some shares for $1000. Each day, she has to decide whether to sell, or keep them. Her profit will depend on the price she gets for them. Each morning, she gets the previous day’s closing price, and she rings her sister Megan and gets her opinion as to whether the price will go up or
Peter wants to do simple monitoring of his local weather. Suppose that the main weather types are: sunny, cloudy, showers, rain and snow. He knows that the weather changes from one day to the next in certain patterns, depending on the season. He also has access to the local weather forecast for the
Julia’s manufacturing company has to decide whether to go ahead with the production of a new product. Her analysis indicates that the future profit will depend on a combination of the quality of the product and the market demand for it. Before the final decision is made, she has two other
Think of your own decision making problem involving reasoning with evidence and uncertainty. Write down a text description of the problem, then model it with a decision network using similar steps to the previous problem. Make the problem sufficiently complex that your network has at least 3 chance
Robert is trying to decide whether to study hard for the Bayesian Artificial Intelligence exam. He would be happy with a good mark (e.g., a High Distinction) for the subject, but he knows that his mark will depend not only on how hard he studies but also on how hard the exam is and how well he did
You have 20 utiles worth of chocolate. We offer to make a gamble with you: tossing some coin, if it comes up heads, we’ll give you 130 more utiles worth of chocolate;if it comes up tails, we’ll take your 20 utiles worth of chocolate. Assuming that you maximize your expected utility and that you
You have 20 pounds of cheese. We offer to make a gamble with you: tossing a fair coin, if it comes up heads, we’ll give you 130 more pounds of cheese; if it comes up tails, we’ll take your 20 pounds of cheese. What is the expected value of the gamble?Assuming that you maximize your expected
Take Pearl’s earthquake example. Suppose there is an intervention on JohnCalls.1. Use ordinary observation and updating. What is Bel(Burglary)?2. Use Pearl’s cut-link method. What is Bel(Burglary)? this case?3. Use the augmented model with 0.9 effectiveness. What is Bel(Burglary)?Now add an
As mentioned in the Bibliographic Notes, Hugin implements an approximation scheme that involves setting very small probabilities in the junction tree tables to zero, which in turns allows more compression to take place.To turn on the approximation scheme, select Network!Network
The on-line material for this text provides a version of both the Logic Sampling algorithm (Algorithm 3.3) and the Likelihood Weighting algorithm (Algorithm 3.4).Take an example BN (either provided with the online material, or that you’ve developed for the problems set in Chapter 2) and do the
Consider the BNs for the “Asia” problem described in x2.5.3.1. Use the Jensen junction tree algorithm (Algorithm 3.2) to construct a junction tree from both these networks, drawing (by hand) the resultant junction trees.2. Load these networks (asia1.dne and asia2.dne in the on-line
.8, Chapter 2) the belief network for another version of the medical diagnosis example, where B=Bronchitis, S=Smoker, C=Cough, X=Positive X-ray and L=Lung cancer and all nodes are Booleans. Suppose that the prior for a patient being a smoker is 0.25, and the prior for the patient having
Consider (again — see
Modify your network construction algorithm from Problem 9 above to use the dseparation oracle from the last problem, instead of input from the user. Your new algorithm should produce exactly the same network as that used by the oracle whenever the variable ordering provided it is compatible with
Given the internal Bayesian network structure N, implement a d-separation oracle which, for any three sets of nodes input to it, X, Y, and Z, returns: true if X ?YjZ (i.e., Z d-separates X and Y in N); false if X 6? YjZ (i.e., X and Y given Z are d-connected in N); some diagnostic (a value other
Given as input the internal Bayesian network structure N (in the representation you have designed above), write a function which returns all undirected paths (Definition 2.1) between two sets X and Y of nodes in N.Test your algorithm on various networks, including at least The d-separation network
Implement the network construction algorithm (Algorithm 2.1). Your program should take as input an ordered list of variables and prompt for additional input from the keyboard about the conditional independence of variables as required. It should generate a Bayesian network in the internal
Design an internal representation for a Bayesian network structure; that is, a representation for the nodes and arcs of a Bayesian network (but not necessarily the parameters — prior probabilities and conditional probability tables). Implement a function which generates such a data structure from
Consider the following graph.1. Find all the sets of nodes that d-separate X and Y (not including either X or Y in such sets).2. Try to come up with a real-world scenario that might be modeled with such a network structure. X A B Y
Consider the Bayesian network given for the previous problem.1. What variable ordering(s) could have been used to produce the above network using the network construction algorithm (Algorithm 2.1)?2. Given different variable orderings, what network structure would result from this algorithm? Use
Consider the following Bayesian network for another version of the medical diagnosis example, where B=Bronchitis, S=Smoker, C=Cough, X=Positive X-ray and L=Lung cancer and all nodes are Booleans.List the pairs of nodes that are conditionally independent in the following situations:1. There is no
Come up with your own problem involving reasoning with evidence and uncertainty.Write down a text description of the problem, then model it using a Bayesian network.Make the problem sufficiently complex that your network has at least 8 nodes and is multiply-connected (i.e., not a tree or a
“Jason the Juggler.” Jason, the robot juggler, drops balls quite often when its battery is low. In previous trials, it has been determined that when its battery is low it will drop the ball 9 times out of 10. On the other hand when its battery is not low, the chance that it drops a ball is much
“A Lecturer’s Life.” Dr. Ann Nicholson spends 60% of her work time in her office.The rest of her work time is spent elsewhere. When Ann is in her office, half the time her light is off (when she is trying to hide from students and get research done). When she is not in her office, she leaves
“Fred’s LISP dilemma.” Fred is debugging a LISP program. He just typed an expression to the LISP interpreter and now it will not respond to any further typing.He can’t see the visual prompt that usually indicates the interpreter is waiting for further input. As far as Fred knows, there are
Construct a network in which explaining away operates, for example, incorporating multiple diseases sharing a symptom. Operate and demonstrate the effect of explaining away. Must one cause explain away the other? Or, can the network be parameterized so that this doesn’t happen?
A DNA match between the defendant and a crime scene blood sample has a probability of 1/100000 if the defendant is innocent. There is no other significant evidence.1. What is the probability of guilt?2. Suppose we agree that the prior probability of guilt under the (unspecified)circumstances is
(Note: this is the case of Sally Clark, convicted in the UK in 1999, found innocent on appeal in 2003, and tragically died in 2007 of alcohol poisoning. See Innocent, 2002.)A mother was arrested after her second baby died a few months old, apparently of sudden infant death syndrome (SIDS), exactly
You are offered a Keno ticket in a casino which will pay you $1 million if you win!It only costs you $1 to buy the ticket. You choose 4 numbers out of a 9x9 grid of distinct numbers. You win if all of your 4 numbers come up in a random draw of four from the 81 numbers. What is the expected dollar
A bookie offers you a ticket for $5.00 which pays $6.00 if Manchester United beats Arsenal and nothing otherwise. What are the odds being offered? To what probability of Manchester United winning does that correspond?
Construct a Dutch book against someone who violates the Axiom of Additivity.That is, suppose a Mr. Fuzzy declares about the weather tomorrow that P(Sunny) =0:5;P(Inclement)=0:5; and P(Sunny or inclement)=0:5. Mr. Fuzzy and you agree about what will count as sunny and as inclement weather and you
(Adapted from Fischoff and Bar-Hillel (1984).) Two cab companies, the Blue and the Green, operate in a given city. Eighty-five percent of the cabs in the city are Blue;the remaining 15% are Green. A cab was involved in a hit-and-run accident at night.A witness identified the cab as a Green cab. The
You consider the probability that a coin is double-headed to be 0.01 (call this option h0); if it isn’t double-headed, then it’s a fair coin (call this option h). For whatever reason, you can only test the coin by flipping it and examining the coin (i.e., you can’t simply examine both sides
After winning a race, an Olympic runner is tested for the presence of steroids. The test comes up positive, and the athlete is accused of doping. Suppose it is known that 5% of all victorious Olympic runners do use performance-enhancing drugs. For this particular test, the probability of a positive
The frequency of male births at the Royal Women’s Hospital is about 51 in 100. On a particular day, the last eight births have been female. The probability that the next birth will be male is:1. About 51%2. Clearly greater than 51%3. Clearly less than 51%4. Almost certain 5. Nearly zero
The probability of a child being a boy (or a girl) is 0.5 (let us suppose). Consider all the families with exactly two children. What is the probability that such a family has two girls given that it has at least one girl?
There are five containers of milk on a shelf; unbeknownst to you, two of them have passed their use-by date. You grab two at random. What’s the probability that neither have passed their use-by date? Suppose someone else has got in just ahead of you, taking one container, after examining the
Prove the theorems of x1.3.1, namely the Total Probability theorem and the Chain Rule.
Given that two pieces of evidence e1 and e2 are conditionally independent given the hypothesis — i.e., P(e1je2;h) = P(e1jh) — prove the “product rule”: P(e1;e2jh) =P(e1jh)P(e2jh).
Prove that the conditional probability function P(je), if well defined, is a probability function (i.e., satisfies the three axioms of Kolmogorov).
=+8. Name prominent CEP products
=+7. How can CEP be used for semantic enrichment?
=+6. Explain filter, pattern matching, value progression analysis and time progression analysis.
=+5. Explain the difference between data access in a DBMS versus a CEP engine?
=+4. What is a CEP engine?
=+3. What is a message broker?
=+2. What is an event, event object and event type?
=+1. What is CEP?
=+6. What are the advantages / disadvantages of using a web service compared to a library or framework?
=+5. Explain transfer learning
=+4. Explain deep learning. What are convolutional layers? What are pooling layers? What are dense layers?
=+3. Which libraries / frameworks / web services can be used for computer vision?
=+2. What are the main tasks and approaches in computer vision?
=+1. Name applications of computer vision
=+• How to generate images / videos from an internal representation, e.g., in a computer game?
=+• What is depicted in an image / which situation takes place in a video, e.g., is there a danger of a vehicle collision?
=+• How to retrieve and classify images and videos, e.g., which photos depict certain people?
=+7. What do NLP web services offer? Give examples.
=+6. What do NLP libraries and pipeline frameworks offer? Give examples.
=+5. What do language resources offer to NLP? Give examples.
=+4. What is tokenization, sentence splitting, PoS tagging, and parsing?
=+3. Explain the bag-of-words model, tf-idf and the n-gram model.

Showing 600 - 700 of 4756

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Last

Applied Artificial Intelligence An Engineering Approach 2nd Edition Bernhard G Humm - Solutions

Step by Step Answers