The final project is an opportunity to apply what you have learned in class to a problem of your interest. It contributes 30% to the final grade.
Suitable research topics for the final project:
AI competitions and challenge problems EvalAI hosts many interesting competitions with prizes suitable for students from all backgrounds, e.g., (i) Open Catalyst Challenge for Chemical Engineering students; (ii) Neural Latents Benchmark for Neuroscience students.
Hackthon of ML and AI: Kaggle, open graph benchmark
Apply a state-of-the-art deep learning or machine learning algorithm to a new problem (your own research project is recommended).
Improve an existing data mining/machine learning algorithm.
Performance evaluation of several existing ML algorithms.
What is the problem that you will be investigating? Why is it interesting?
What data will you use? Give examples of the dataset.
What data mining or machine learning method or algorithm are you proposing? If there are existing implementations, will you use them, and how? How do you plan to improve or modify such implementations? You don’t have to have an exact answer at this point, but you should have a general sense of how you will approach the problem you are working on.
How will you evaluate your results? Qualitatively, what kind of results do you expect (e.g., plots or figures)? Quantitatively, what kind of analysis will you use to evaluate and/or compare your results (e.g., what performance metrics or statistical tests)?
Submission: Please submit your proposal as a PDF on the college dropbox. Only one person on your team should submit. Please have this person add the rest of your team to the author list. (Due in the week after Fall break).
Each group needs to write a project report of at least 5 pages, not including references.
The project report should be formatted similarly to a workshop paper.
The suggested structure:
Title, Author(s)
Abstract: Briefly describe your problem, approach, and key results.
Introduction (10%): Describe the problem you are working on, why it’s important, and an overview of your results.
Related Work (10%): Discuss published work related to your project. How is your approach similar or different from others?
Data (10%): Describe the data you are working with for your project. What type of data is it? Where did it come from? How much data are you working with? Did you have to preprocess, filter, or other special treatments to use this data in your project?
Methods (25%): Discuss your approach to solving the problems you set up in the introduction. Why is your approach the right thing to do? Did you consider alternative approaches? You should include figures, diagrams, or tables to describe your method.
Experiments (40%): The exact experiments will vary depending on the project, but you might compare with previously published methods, use visualization techniques to gain insight into how your model works, discuss standard failure modes of your choices. You should include graphs, tables, or other figures to illustrate your experimental results.
Conclusion (5%): Summarize your key results - what have you learned? Suggest ideas for future extensions or new applications of your thoughts.
Supplementary Material, not counted toward your 6-8 page limit and submitted as a separate file
Declaration of contributions of all group members.
Your supplementary material might include:
Source code
IPython notebook, videos, interactive visualizations, demos, etc..
Some examples of good project reports:
Scheduled on the final exam day for this course.
your presentation should include the following sections. (10 minutes each group, 12-15 slides)
Problem Statement: Briefly describe the problem your group is tackling. Describe the overall motivation and the input/output of the problem.
Technical Challenges: Briefly describe why the problem is technically challenging.
Related Works: Briefly in what ways previous works have tackled the technical challenges.
Your Approach and Results: Describe your detailed technical approach and innovations. Describe evaluation results (dataset and metric). Emphasize important, interesting, or unexpected results and explain the expected results compared with previously reported results.
Broader Impact: How do you expect the impact of your work to be? What can others learn from it, or how can they apply it to solve their problems? What are the limitations of your work? What are areas for future improvements?
Project Report: in PDF format.
Code: put the GitHub or GitLab link in your report. Please make sure the repository is public, and you put all materials, including all code, scripts, tutorials, everything that one requires to replicate the results in your report.
Data: Including training dataset, testing dataset, metadata, and more importantly, all experimental data, logs that you have collected during your experiments, everything that one requires to replicate the results.
Project Presentation ppt: you can use Keynote or PowerPoint to prepare the slides and convert them to PDF. You can publish your presentation on SlideShare/SpeakerDeck and put the link in your report. If you prefer, you can push the PDF into your git repository.
Upload all materials as a zip file to http://dropbox.cse.sc.edu