Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
CS3910 Computational Intelligence Lab
The intention is to check your proficiency in implementing the taught computational intelligence methods on the two example problems studied in the practical tasks: 1. The TSP problem, including a correct cost/fitness function, 2. Random search implemented on the TSP, 3. Local search implemented on the TSP, 4. One computational intelligence algorithm of your choice implemented on the TSP, 5. One computational intelligence algorithm of your choice implemented on the antenna array design problem. Descriptive details of Assignment: There will be 5 tasks that you will be asked to demonstrate. At the start of the lab sign-off you will be provided with a new, unseen csv file, formatted in the same way as ulysses16.csv and you will be asked to use this for demonstrating your TSP solution. The tasks you will be asked to perform are: 1. Show the calculation of the cost function for a route in the TSP that is given to you in the demo (based on Practical 1). For example, if the csv file contained 10 cities, you may be asked to show the cost of route 1-4-2-3-6-5-7-8-10-9-1. You may make the assumptions that all routes start and finish at the first city. 2. Show your implementation of random search on TSP (based on Practical 1). This should print out the best routes found by random search together with their costs. You may use a termination condition based on time (for example 30 seconds) or a pre-set number of tries (for example 100). 3. Show your implementation of local search for TSP (based on Practical 2). This should print out the best routes found by local search together with their costs. 4. Show your implementation of a computational intelligence algorithm of your choice implemented on the TSP (based on Practical 5). This can be a genetic algorithm with swap mutation and order 1 crossover, operating on a small budget of population size (for example 50) and number of generations (for example 50) to allow for the algorithm to be demonstrated in one minute. 5. Show your implementation of a computational intelligence algorithm of your choice implemented on the antenna array design problem. You will be provided with a number of antennae and an angle as values to use in the sign-off. Please be prepared to answer a small number of questions about your code and to show sections of your code where particular functionalities are implemented. Available on Blackboard is a test.csv file, in the same format as the ulysses16.csv file, for the purpose of testing your own TSP solution. Note that, just like ulysses16.csv, this test file contains two header rows and city numbers in the first column. The file you will be asked to use in the sign off will be of exactly this format, so it is important that your code can handle these features without you having to manually edit the csv file. Please note that there will not be any time for debugging or discussing unexpected behaviour of your code during the sign-off. Therefore, it is essential that you practice showing your code and your results before the demo, to ensure it runs smoothly. Please book a 5 minutes appointment for the lab sign-off during the stated period as instructed in the Blackboard announcement and email that will be sent during week starting 1st November 2021 (the week before the start of the lab sign-off assessment). In preparation for the sign-off, submit your source code as a single doc, pdf or rtf file by the stated deadline. Should you need any feedback on your implementation after the sign-off session, feel free to book an appointment with either of the tutors during their respective office hours. Recommended reading/ online sources: Agoston E Eiben, James E Smith (2015). Introduction to evolutionary computing. Springer On Blackboard, you will find all relevant lecture material in the Lecture material folder within Learning Resources and all relevant lab material in the Lab material folder within Learning Resources. Key Dates: 22/10/2021 Experimental study: sign-off released 29/10/2021 Experimental study: full task released 08/11/2021 Submission of code for lab sign-off 10/11/2021-18/11/2021 Sign-off session, to be booked by each student as a single individual WASS appointment of 5 minutes 26/11/2021 Lab sign-off marks provided to students Submission Details: Sign-off: book session as described above (please monitor Blackboard announcements and emails about this) Lab sign-off code: Online on Blackboard Marking Rubric: Please note that the lab sign-off contributes to 10 marks in the Experimental Study. For each task 0,1, or 2 marks will be allocated as follows: 0 for a task: For the given task, the implementation does not behave as expected. 1 for a task: For the given task, the implementation shows behaviour close to that expected, with minor flaws only. 2 for a task: For the given task, the implementation behaves as expected.