Scribd
Upload
114 views2 pages

AI Final Project Instruction

This document provides instructions for the final project in the Introduction to Artificial Intelligence course. Students must choose a problem to solve, implement two different solutions to the problem, write a report analyzing the solutions, and present their work and solutions. The project will be graded based on various aspects of problem selection, solution modeling and implementation, analysis of results, reporting, and presentation. Students are encouraged to seek support from course staff and should contact them as early as possible if needing help.

Uploaded by

尹鑫圳
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
114 views2 pages

AI Final Project Instruction

This document provides instructions for the final project in the Introduction to Artificial Intelligence course. Students must choose a problem to solve, implement two different solutions to the problem, write a report analyzing the solutions, and present their work and solutions. The project will be graded based on various aspects of problem selection, solution modeling and implementation, analysis of results, reporting, and presentation. Students are encouraged to seek support from course staff and should contact them as early as possible if needing help.

Uploaded by

尹鑫圳
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

COURSE 67842 INTRODUCTION TO ARTIFICIAL INTELLIGENCE

Final project instructions

June 6, 2021

1 P ROJECT D ESCRIPTION
In your projects, you will use the ideas and techniques you learned in class in order to solve a
new problem on your own.
First, choose a problem. Define and describe it, explain why it is not trivial and why it is inter-
esting.
Second, implement two different ways to solve your problem. You are encouraged to go over
course material and decide what chapters are relevant to your project. What are the assump-
tions on your problem’s world? What algorithms best fits your needs? What algorithms has no
chance of working on this setup? Implementation may use external libraries, code snippets or
be based on your previous exercises, but you must properly reference all imported code frag-
ments and cite the original authors.
Third, write a report. Measure your solutions’ performances and present the results. Explain
the differences between solutions and compare it with trivial bounds.
Forth, Present your work. Make a short story (a video presentation, about 6-8 minutes long)
presenting your project to your colleagues. Try to separate the wheat from the chaff - supply
motivation to your work and talk about the important details and results, but omit unimportant
technicalities.
Fifth, Peer review. Ask a meaningful question on others’ presentations. Exact matching and
further instructions will be published after project submission.

2 S CHEDULE
Make sure you are preforming the following steps before their deadlines (Remember: sooner is
better!). These dates are tentative and might change.
ASAP - Sign as a group on the course Moodle.
11/7 - Hand out your projects proposal, as described on the projects recitation.
31/8 - submit your final project.
7/9 - finish peer review task.

1
3 G RADING
Each projects will be checked and evaluated according to the following aspects:

Choosing a Problem (5%): Did you choose a reasonable problem? Is it hard enough so classic
programming techniques won’t be able to solve it too easily?
Solutions (15%): Do your solution ideas make sense? Did they have any chance of working in
the first place?
Modeling (10%): Problems are highly detailed and very down-to-earth creatures. Algorithms
are abstract and generalized ideas. Have you managed to connect them without breaking any-
thing? Were the essential parts of both the algorithm and the problems taken in to account in
your project? Were any details you neglected more than just minor details?
Working algorithm (5%): It should work (also for us! include instructions for running your
code).
Performance Measurements (5%): You measured some properties of your results. Are these
measurements meaningful? Are they random? Are they too coarse to reveal meaningful differ-
ences between different runs?
Presenting results (10%): Did you present your results in a clear and a reasonable way? Can
we look at your graphs and understand what actually happened?
Result analysis (10%): Are your conclusions consistent with the results you got? Were you able
to criticize your own work? Have you noticed weird anomalies or could you spot mistakes, and
can you explain what is going on?
Impressive result (5%): Have you managed to actually solve the problem you were working on?
Did you get non-trivial results? Can your solution beat you/state-of-the-art/standard methods
on this specific setup?
Report (15%): Is your report well-written? Does it contradict itself?
Story (15%): Except for being cool and catchy, does your story presents the essentials of your
work? Is watching it is generally a positive experience?
Peer review (5%): Could you ask *meaningful* *questions* about your colleagues works?
Bonus (5%): If we noticed something really problematic or pretty awesome that was not graded
in any other way, you will get it here

The projects final grades will be a normalized version of the individual evaluations,
according to the Faculta’s instructions.

4 G ET SUPPORT
If you are stuck or unsure about any part of the project- please contact the course staff. Extra
office hours will be given on the project period. You are more than welcome to consult with us,
and we will do our best to direct and guide you on this task.
Please contact us as soon as you discover a problem, it might take a few days until we are free
to meet you, so don’t wait for the last minute.

We are exited to see your projects! Good luck!

You might also like