Games and Artificial Intelligence Techniques COSC 2527/2528 Assignment 2 Assessment Type Group assignment. Groups as allocated and notified on Can- vas. Submit online via Canvas → Assignments → Assign- ment 2. Marks awarded for meeting requirements as closely as possible. Clarifications/updates may be made via announce- ments/relevant discussion forums. Due Date 11th October 2020, 11:59pm Marks 40 1 Overview In this assignment your group is expected to demonstrate a deep understanding of an advanced AI technique, and how to implement it effectively in the context of a game. The technique of choice should be discussed with your instructors during the weekly discussions in class. Your instructors will provide weekly feedback during the inclass progress discussions. 2 Learning Outcomes This assessment relates to all of the learning outcomes of the course which are: • Demonstrate an understanding of various AI techniques and tools and how they are applied in the context of games programming; • Design and develop a gaming application, based on existing games engines or plat- forms; • Work effectively in a team environment to develop a complex software system. 3 Assessment details Working with your last assignment as a starting point you must, in your groups, extend your game to use an advanced AI learning technique of your choice. Possible choices include neural networks or decision trees, with either online or offline learning. If you wish to try something other than these (such as agent oriented programming), please discuss this with the course instructors. How you integrate this into your game is entirely up to you. As in the last assignment you are encouraged to be creative, and more marks will be given to creative solutions. Your game rules will probably have to be altered or extended for your new AI imple- mentation. While a more complex game environment justifies a more complex solution, you may wish to disable some components of your first assignment in order to make the AI implementation tractable. For example, if you intend to use reinforcement learning, you may wish to disable pathfinding so that you can speed up the game during train- ing. Regardless of any decisions you make here, your game should be able to effectively demonstrate the strengths of your new agents. You are permitted to use external libraries, and there will be no mark penalty for doing so as long as you reference them and are able to explain the algorithms and processes clearly. However coding your own solution is strongly encouraged. For groups whose first assignment was incomplete or otherwise difficult to use as a basis for Assignment 2, we will provide a small starter game that you can use as a template. You will not receive any grade penalty if you choose to use this template. 4 Submission You have to submit all the relevant material as listed below via Canvas. A group will be available for your team, so any of the members can upload the assignment. 4.1 Submission Components Implementation: Please submit the source code for your implementation, not your compiled game. We do not mind which version of Unity you use, provided that it is recent (2019 or later) and all members of your team use the same version. Report: A full report is required with submission. As with the first assignment, we will provide you with a report template, although you are not bound to it. Regardless, the report must cover the following: • Design and Intention: A brief summary of the gameplay of your new design. What have you changed from your previous assignment and why? Talk about the behaviours you intended to create for your agents. • Algorithm choice: Explain and justify your algorithm choice in terms of your game - how can it achieve the behaviour you intend? • Implementation: Document the process of implementing and tweaking your AI in detail. Talk about changes you had to make. Describe any problems you en- countered, and how they were overcome. • Outcome: How the behaviour you managed to achieve compares to your origi- nal predictions. Were you able to achieve the behaviour you wanted? Was any unexpected behaviour observed? Although each team will submit a single project and report, marks will be awarded individually. Subsequently you must also include a summary of each team member’s contributions noting which group member(s) were responsible for which parts of the im- plementation. Presentation: Your group will be required to record a video presentation of your project. The presentation should be an accessible, high-level summary of your work and is an opportunity to show off your project to the class. Presentations are to be 15 minutes in duration and each team member should present for roughly an equal amount of time. 2 4.2 Submission format Please submit the following files via Canvas. Only one person from each group should perform the upload to avoid versioning issues. Also, please check all elements of your assignment after submission to ensure that they can be opened. • Implementation: Single zip file containing the source code for your assignment. We strongly recommend that you include a README file with instructions on how to run your game. If your implementation involves any kind of machine learning algorithm, we recommend that you include separate options for training a new model from scratch, and loading a pre-trained model from a save file. Make sure that the game can run with only the code included in your zip file! • Report: Single pdf file. • Presentation: A video file in any suitable format. Please note that this file will be shared with your colleagues. 4.3 Late Penalty After the due date, you will have 5 business days to submit your assignment as a late submission. Late submissions will incur a penalty of 10% per day. After these five days, Canvas will be closed and you will lose ALL the assignment marks. 4.4 Assessment declaration When you submit work electronically, you agree to the assessment declaration: https:// www.rmit.edu.au/students/student-essentials/assessment-and-exams/assessment/ assessment-declaration 5 Teams Team membership is as the first assignment. Exceptions need explicit approval from the course coordinator. Marks are awarded to the individual team members, according to the contributions made towards the project. As in the first assignment, we ask that you please help us in this regard by making your team’s inner workings as visible as possi- ble. Communication should be handled via your assignment group channel in Microsoft Teams, and your code repo should be shared with the instructors. Every week during the workshop/lab sessions, you will be required to discuss your weekly progress with respect to this assignment with the teaching staff. These discussions will inform the final assessment of your submissions, so it is important for all group members to attend. 6 Academic integrity and plagiarism (standard warning) Academic integrity is about honest presentation of your academic work. It means ac- knowledging the work of others while developing your own insights, knowledge and ideas. You should take extreme care that you have: 3 • Acknowledged words, data, diagrams, models, frameworks and/or ideas of others you have quoted (i.e. directly copied), summarised, paraphrased, discussed or men- tioned in your assessment through the appropriate referencing methods • Provided a reference list of the publication details so your reader can locate the source if necessary. This includes material taken from Internet sites. If you do not acknowledge the sources of your material, you may be accused of plagiarism because you have passed off the work and ideas of another person without appropriate referencing, as if they were your own. RMIT University treats plagiarism as a very serious offence constituting misconduct. Plagiarism covers a variety of inappropriate behaviours, including: • Failure to properly document a source • Copyright material from the internet or databases • Collusion between students For further information on our policies and procedures, please refer to the following: https://www.rmit.edu.au/students/student-essentials/rights-and-responsibilities/ academic-integrity. 7 Marking guidelines • Implementation 17/40; • Report 15/40; • Presentation 8/40 4
欢迎咨询51作业君