Grading Questions		Grading Scale
Criteria	Suggested Prompts	Poor (1 pt)	Fair (2 pts)	Good (3 pts)	Very good (4 pts)	Excellent (5 pts)
Agent Reasoning	What is your agent's reasoning approach - visual or verbal? Does your agent need to convert between one or the other?	No clear response or incorrect response. Shows a lack of understanding	Provides answer without enough detail. No justifications provided	Provides answers with enough details and examples, but no adequate justifications	Author provides answer with more details about the agent's process. Gives examples and justifications as well	In addition to the criteria for "Very Good", the author added insightful and novel interpretation that expands on the concepts taught in class.
Agent's data representation and problem solving	How does your agent represent/store the images efficiently? What is your agent’s overall problem-solving process?	No clear response or incorrect response given. Shows a lack of understanding	Provides answer to either data representation or problem solving. No clear connection of how they both work together to solve RPMs is given	Provides a clear succinct answer addressing both data representation and problem solving. Answer makes a clear connection on how they work together to solve RPMs	Provides answer to a "good" criteria and makes a strong connection on how they work together, and is able to relate them to human cognition	In addition to the criteria for "Very Good", the author added insightful and novel interpretation that expands on the concepts taught in class.
Agent design	Provide an overview of the design of your agent	Is not able to identify the different components or how they work together	Identifies components but does not provide details on how they work together.	Identifies the various parts and components, and is able to provide a detail explanation on how they work together.	Provides the high level design with the various components in enough detail. The justifications/explanations are also given	Author provides a detailed high level design with design justifications. Draws parallels to design patterns/systems from the real world. Adding other novel interpretations are highly encouraged.
Agent limitations	What mistakes does your agent makes? Could these mistakes be resolved within your agent’s current approach, or are they fundamental problems with the way your agent approaches these problems?	Does not provide an answer or has a very poor/ill structured answer	Is able to identify mistakes but does not provide details or explanations	Identifies mistakes, provides analysis and explanations with examples and details	Identifies mistakes, provides analysis and explanations with examples and details. Provides some ideas on how they intend to solve them.	In addition to the criteria for "Very Good",author gives more detailed ideas on how they intend to solve the mistakes. The author's interpretation for the fundamental reason for the mistakes is very novel
Agent performance	Please detail on your evaluation/performance criterias and your agent's results. Think about accuracy, efficiency and generality. Are there other metrics or scenarios under which you think your agent's performance would improve or suffer?	Does not provide an answer or has a very poor/ill structured answer	Provides only the agent scores and no other insights	In addition to scores, details one or more of their own evaluations metrics and gives details on how their agent performs	Provides scores, and evaluation metrics apart from the ones mentioned in the rubric. Provides a detailed performance overview of the agent	In addition to the criteria for "Very Good", the author provides justifications/explanations for the agent's performance.
Agent's relation to KBAI	Please provide an explanation of how your methods/components/ideas in your agent's design are/might be similar to (or can be related) to specific KBAI methods discussed in class	Does not provide an answer or has a very poor/ill structured answer	Is able to provide one or more KBAI concepts but does not adequately relate them	Is able to provide one or more KBAI concepts with adequate explanation on how they relate	Discusses one or more KBAI concepts, shows how they relate to their agent design and provides enough justification of why using that KBAI concept over the other was a good idea	In addition to the criteria for "Very Good", the author added insightful and novel interpretation that expands on the concepts taught in class.
Agent's relation to human cognition	What does the design and performance of your agent tell us about human cognition? How is it similar, and how is it different? Has your agent’s performance given you any insights into the way people solve these problems?	No answer provided.	Gives some/partial explanation of how the agent can be related to human's reasoning process but cannot give enough detail to back claims	Provides analogies between agent design and human cognitions with explanations	In addition to providing relation to human cognition, the author gives detailed examples and is able to justify the performance	In addition to the criteria for "Very Good", the author goes above and beyond to provide explanations about the failures/mistakes and relates them to human reasoning failures if any. Additional novel interpretations and explanations would fetch you full marks
Report	Points for effective report - readable, concise, organised, citations	Report doesn't cite sources, overshoots the recommended word limit, and does not effectively use diagrams to make report concise	No sources cited but the report remains within word limit	Report is within word limit and uses recommended format for citing sources	Report uses diagrams/infographics to give concise explanations. Is within resonable word limit and cites sources.	In addition to the criteria for "Very Good",the author uses infographics/diagrams/flowcharts very effectively