Course Project

The Course project is an opportunity for you to delve deep into any Machine Learning problem or topic of your choice. Your project could be of a theoretical and/or applied nature. Ideally we expect you to identify an interesting problem, develop new methodolgy, analyse the properties of such methods and apply it on real data. However, the main focus could be on just one or two of them. Since most taking this course are advanced graduates we expect you to aim for something publishable.


Your project will be worth 40% of your final class grade, and will have 5 deliverables:


  1. Proposal : 2 pages (10%)
    Due : October 8
  2. Midway Report : 4-5 pages (20%)
    Due : November 5
  3. Final Report : 8 pages, 9 pages including references (35%)
    Due : Dec 3
  4. Presentation : (25%)
    On : Dec 1, 4.30 pm - 7.30 pm GHC 4215
  5. Peer Review : (10%)

Note: All reports should be in NIPS format.

Project Proposal

You must turn in a brief project proposal that provides an overview of your idea and also contains a brief survey of related work on the topic. The proposal should also include a plan of activities, including what you plan to complete by the midway report. Each team should submit a hard copy and email a soft copy to 10715-instructors@cs before class.

Midway Report

You must have a solid introduction and literature survey in the midway report. If you are proposing a new method, then include the details of the method and a plan for experiments. If you are working on a theory project, include the final results you want to prove and any preliminary results.

Final Report

The final report should at most 8 pages long. You can use one extra page for references. The report should have an introduction motivating the problem you are solving and discussing related prior work. You should detail your methodological and/or theoretical contributions to solve the problem and experimental results. You may include any supplementary material with your submission in addition to the 8 page report. However, you should assume that you will be graded solely based on your report.

Peer Review

During the presentation session, you will also be required to evaluate the projects of other students. We will send you a guideline shortly before the deadline. You will be graded based on your evaluations of other projects.


Here are the projects for this semester.

Information Theoretic Clustering via Kernel Density Estimation
Shashank Singh, Bryan Hooi

Visual Knowledge Discovery Using Deep Learning
Gunnar Atli Sigurdsson, Shi Hu

Modified Dropout for Training Neural Networks
James Duyck, Min Hyung Lee, Eric Lei

Sparse Supervised Topic Model
Yining Wang, Chun-Liang Li, Kevin Lin

Predicting Latent Attributes of Twitter Users
Yotam Hechtlinger, Natalie Klein, Hyun Ah Song

How Many Random Restarts ?
Christoph Dann, Travis Dick, Eric Wong

Context Effects in Sentence Reading
Jonathan Mei, Mariya Toneva

Using Multi-task Learning to Predict Signalling in Regulatory Pathways
Krishna Pillutla, Rohan Varma, Petar Stojanov

Canonical Correlation Analysis and Graphical Modeling for Huaman Trafficking Characterization
Qicong Chen, Maria De Arteaga, William Herlands

Distributed Inference of Overlapping Communities
Xun Zheng, Jingwei Zhuo

Creating Scalable and Interactive Web Applications Using High Performance Latent Variable Models
Aaron Li, Yuntian Deng, Kublai Jing, Joseph Robinson

© 2014 Eric Xing @ School of Computer Science, Carnegie Mellon University
[validate xhtml]