CS5824/CS4824/ECE5424/ECE4424: Machine Learning, Fall 2015 Project Page
Project Presentation Schedule
Groups will be presenting 12 minute talks on their projects in Durham 261. The tentative schedule is below, with authors listed in alphabetical order. Visitors are very welcome to attend these class sessions.
Tuesday, December 1.
- 2:00 pm: Christopher Cronin, Connor Sullivan.
Improving Elo Ranking for Sports
Download paper
- 2:15 pm: Miraziz Yusopov.
k-Means Approach to Legislative Redistricting
Download paper
- 2:30 pm: Andrew Ciambrone, Arpit Goyal, Hossameldin Shahin.
Hand Gesture Recognition with Batch and Reinforcement Learning
Download paper
- 2:45 pm: Bijaya Adhikari, Jason Granstedt, Aroma Mahendru, Arijit Ray
Learning to Listen: Matching Song Covers to Original Songs via Supervised Learning Methods
Download paper
- 3:00 pm: Walid Chaabene, Sirui Yao.
Enhancing Matrix Factorization Based Recommender Systems Using Co-purchase Networks
Download paper
Thursday, December 3.
- 2:00 pm: Mehda Baidya, Samuel Maddock, Sazzadur Rahaman, Jason Ziglar.
Hierarchical Clustering for Semi-Supervised Ground Truth Generation
Download paper
- 2:15 pm: Saurabh Chakravarty, Sina Dabiri, Krati Nayyar, Saket Vishwarao.
Anomaly Detection for Univariate Time-Series Data
Download paper
- 2:30 pm: Jinwoo Choi, Siddarth Narayanan (in ECE 5554).
Visual Question Answering with Various Feature Combinations: Extensions of Visual Question Answering
Download paper
- 2:45 pm: Sneha Mehta, Aditya Pratapa, Philip Summers.
Birds in a Forest
Download paper
- 3:00 pm: William Doan, Griffin Jarmin.
Simulated Transfer Learning Through Deep Reinforcement Learning
Download paper
Slides are due at noon the day of your talk. Send early drafts if you want to test compatibility.
Overview
The final project for the Machine Learning course will be a mini research project done by groups of 2-4 people. The overall goal of such a project is to do something that nobody else has ever done before. Ways to accomplish this goal include (but are not limited to):
- taking an existing machine learning model or algorithm and tweaking it to improve it in some way (make it faster, make it more accurate, etc.);
- taking an existing machine learning model or algorithm and applying it to a brand new application;
- inventing a new machine learning model or algorithm for a new application;
- and deriving a new theoretical analysis for a new or existing machine learning algorithm.
Because this is a class project on a tight schedule, it's important to keep your proposed ideas tractable, but at the same time the motivation of your project should be toward making a significant scientific contribution, one that we can imagine being a published paper if you had enough time to work on it. For the class project, the expectation is you will most likely only have preliminary results.
You will not need to submit code for the project, nor are you required to use any particular programming language, platform, or software. You are free to use whatever helps you accomplish your research goals the best.
Deliverables
The first deliverable is a project proposal that is due in conjunction with Homework 4 (on October 31).
- Your project proposal should be 1-2 pages long.
- It should discuss what problem you plan to address, your preliminary plan for how you will address it, including what algorithms you will use, ideas for how to change the algorithms, and how you will evaluate success.
- It should list at least three recent, relevant papers you and your team will read and understand as background for your contribution. You probably want to have read at least one of these before writing the proposal
- It should discuss a proposed plan of who on your team will be responsible for what tasks. These contributions should just be a prediction for planning purposes, and you are not obligated to adhere to them precisely.
The main deliverable for the class project is a paper. The paper must be in ICML format. (See the format guidelines and LaTeX style files at http://icml.cc/2015/?page_id=151. Use the formatting for camera-ready papers so your names are not anonymized.) The paper must be no longer than 8 pages, not counting the bibliography (you have unlimited space for references).
The final paper should have the usual structure of a scientific paper:
- The paper should begin with an abstract, which should be no longer than two paragraphs describing what problem you are addressing and what your discovories were.
- The paper should have an opening introduction section that describes the motivation, the problem you are addressing, and your approach to solving the problem.
- The middle of the paper should describe your technical contribution in detail. The middle may be broken up into multiple sections, for example one describing an algorithmic approach and another describing experimental setup and results. A reader should be able to reproduce your experimental results or analysis from reading this middle of the paper.
- Your paper must include a section that discusses prior work: research that studies related problems, or the same problem you are addressing, other approaches that are related to your technical approach that may or may not have been applied to the same type of application, or foundational mathematical or scientific ideas you are building on. Establish in this section why none of these existing studies solves the problem your contribution aims to fix.
- The paper should end with a conclusion that summarizes your contributions and discusses open problems that remain.
- Lastly, your paper should end with a list of references. We strongly recommend using bibTeX (or bibLaTeX) to make bibiographies, because it does a lot of formatting automatically.
As part of the final project report, you will need to describe each person's contributions to the project.
The final project report is due 11:59 pm on Tuesday 12/8.
Project Ideas
This section is under construction. You are welcome to propose any idea you want, and you are welcome to include work you are doing in your job, your own research, or your other classes, as long as it is okay with your other professors or supervisors. The list below is a brainstormed list of ideas that could be interesting, but you are not at all obligated to use one of these ideas.