Crowdly
Add to Chrome
Artificial Intelligence Methods (COMP2001 UNUK SPR) (COMP2011 UNUK SPR) (25-26)
Looking for Artificial Intelligence Methods (COMP2001 UNUK SPR) (COMP2011 UNUK SPR) (25-26) test answers and solutions? Browse our comprehensive collection of verified answers for Artificial Intelligence Methods (COMP2001 UNUK SPR) (COMP2011 UNUK SPR) (25-26) at moodle.nottingham.ac.uk.
Get instant access to accurate answers and detailed explanations for your course questions. Our community-driven platform helps students succeed!
Before continuing with the following exercises, confirm that running your implementation of Iterated Local Search with the default experimental configuration yields the following plots. If it does not, check that you have updated your implementation of `DavissBitHillClimbing` to accept all non-worsening moves, if that still doesn't work, you may use the following reference implementation DavissBitHillClimbing.java (download link here). If that still doesn't work, please check your implementation of Iterated Local Search with a peer/friend/lab assistant!
By using your implementation of Iterated Local Search, and appropriately configuring the framework, find a configuration for iIntensityOfMutation (IOM) and iDepthOfSearch (DOS) such that Iterated Local Search performs better than DBHC alone when comparing their median performances for solving MAX-SAT instance 1 with a 10 second computational time budget.
For full context, here is the boxplot produced when running DBHC for 31 trials on MAX-SAT instance 1 given 10 seconds of runtime using the seed of -1414736884 ("hill climbing".hashcode()) from the previous lab exercise:
The results of the previous exercise may have surprised you.Disabling hill-climbing by setting DOS to 0 performs much better than any configuration with hill-climbing enabled. Reflect on the guidelines for designing Iterated Local Search algorithms discussed in lecture 3 and briefly explain in your own words why you think choosing SDHC as the hill-climbing operator is detrimental to the performance of ILS.
The best configuration in terms of obtaining the best median performance was either with IOM=1 and DOS=2, or with IOM=2 and DOS=2. If we create a new ILS algorithm which uses SDHC rather than DBHC, again accepting non-worsening moves, can we expect that one of these configurations also yields the best performance among all other configurations (for IOM and DOS in the range [0,3])?
Now that you have finished implementing Iterated Local Search, look closely at the configuration of IntensityOfMutation and DepthOfSearch as specified in the Lab3ExercisesTestFrameConfig class. Select all of the following statements that we know are definitely true without having to perform any experimental analysis.
Given the box-plots for 3 algorithms A, B and C based on the objective values obtained from 100 trials while solving an instance of a minimisation problem (and no other information), which of the following statement(s) about the search algorithms is definitely true?
jobs to be processed by a single
machine, each job (
j)
with a due
date (
dj) (i.e. hard deadline), processing time (pj), and
a weight (
wj), which one of the following scheduling notations indicate the problem of finding the optimal sequencing of jobs
producing the earliest time for the last job exiting the system (assuming that the time starts at t=0).
Assume that a generic Iterated Local Search (ILS) algorithm is implemented embedding the Improving Only (IO) acceptance method, Davis's Bit Hill Climbing (DBHC) for local search controlled by the depth of search (DOS) parameter, random bit-flip for perturbation controlled by the intensity of mutation (IOM) parameter. These parameters take integer values in the range [0, 10] that correlate to the number of times of calls to DBHC and bit-flip operator before moving on to the next step in the algorithm. For example, DOS=0 (i.e., DOS is set to 0) indicates that DBHC is not applied, or DOS=3 (i.e., DOS is set to 3) indicates that DBHC is applied to a solution for 3 passes over that solution, while IOM=3 (i.e., IOM is set to 3) indicates that 3 bit-flips are applied to the incumbent solution.
Which one of the following options would convert the ILS algorithm into a Random Walk algorithm (randomly sampling the search landscape)?
Assume that a generic Iterated Local Search (ILS) algorithm is implemented embedding the Improving Only (IO) acceptance method, Davis's Bit Hill Climbing (DBHC) for local search controlled by the depth of search (DOS) parameter, random bit-flip for perturbation controlled by the intensity of mutation (IOM) parameter. These parameters take integer values in the range [0, 10] that correlate to the number of times of calls to DBHC and bit-flip operator before moving on to the next step in the algorithm. For example, DOS=0 (i.e., DOS is set to 0) indicates that DBHC is not applied, or DOS=3 (i.e., DOS is set to 3) indicates that DBHC is applied to a solution for 3 passes over that solution, while IOM=3 (i.e., IOM is set to 3) indicates that 3 bit-flips are applied to the incumbent solution.
Is the following statement TRUE or FALSE?
If IOM=0 (IOM is set to 0) and DOS=10 (DOS is set to 10), ILS is likely to get stuck at a local optimum.
Assume that a generic Iterated Local Search (ILS) algorithm is implemented embedding the Improving Only (IO) acceptance method, Davis's Bit Hill Climbing (DBHC) for local search controlled by the depth of search (DOS) parameter, random bit-flip for perturbation controlled by the intensity of mutation (IOM) parameter. These parameters take integer values in the range [0, 10] that correlate to the number of times of calls to DBHC and bit-flip operator before moving on to the next step in the algorithm. For example, DOS=0 (i.e., DOS is set to 0) indicates that DBHC is not applied, or DOS=3 (i.e., DOS is set to 3) indicates that DBHC is applied to a solution for 3 passes over that solution, while IOM=3 (i.e., IOM is set to 3) indicates that 3 bit-flips are applied to the incumbent solution.
Is the following statement TRUE or FALSE?
If IOM=10 (IOM is set to 10) and DOS=0 (DOS is set to 0), ILS becomes a Random Mutation Hill Climbing algorithm.