Machine Learning Fall 2007

10-701 and 15-781

Carlos Guestrin

School of Computer Science, Carnegie Mellon University

About | People | Lectures | Recitations | Homework | Exams | Readings | Projects

• Project Proposal
• Poster Format
• Datasets and Projects
• Tools

Your Course Project
 

Your class project is an opportunity for you to explore an interesting machine learning problem of your choice in the context of a real-world data set.  Below, you will find some project ideas, but the best idea would be to combine machine learning with problems in your own research area. Your class project must be about new things you have done this semester; you can't use results you have developed in previous semesters.

Projects can be done by you as an individual, or in teams of two students.   Each project will also be assigned a 701 instructor as a project consultant/mentor.   They will consult with you on your ideas, but of course the final responsibility to define and execute an interesting piece of work is yours.  Your project will be worth 20% of your final class grade, and will have two final deliverables:

1. a writeup in the format of a NIPS paper (8 pages maximum in NIPS format, including references; this page limit is strict), due December 14th by 2pm (emailed to the instructors list, NO LATE SUBMISSION ACCEPTED since we need to get your grades in), worth 60% of the project grade, and

2. a poster presenting your work for a special ML class poster session on November 30, 2-5pm in the NSH Atrium, worth 20% of the project grade. 

 In addition, you must turn in a midway progress report (5 pages maximum in NIPS format, including references) describing the results of your first experiments by November 12, worth 20% of the project grade. Note that, as with any conference, the page limits are strict! Papers over the limit will not be considered.

 

---

Project Proposal
 

You must turn in a brief project proposal (1-page maximum) by October 17th.  Read the list of available data sets and potential project ideas below.  You are encouraged to use one of these data sets, because we know that they have been successfully used for machine learning in the past.   If you prefer to use a different data set, we will consider your proposal, but you must have access to this data already, and present a clear proposal for what you would do with it. 
 

Project proposal format:  Proposals should be one page maximum.  Include the following information:

• Project title

• Data set

• Project idea.  This should be approximately two paragraphs.

• Software you will need to write.

• Papers to read.  Include 1-3 relevant papers.  You will probably want to read at least one of them before submitting your proposal.

• Teammate: will you have a teammate?  If so, whom?  Maximum team size is two students.

• November 12 milestone: What will you complete by November 12?  Experimental results of some kind are expected here.

 

---

Posters
 

Here are some details on the poster format.

• The two most common ways to make your poster are:
  • You can create a huge slide (in powerpoint, using beamer, etc) and print it out at one of the poster printers available (for SCS, details below). You'd want the shorter dimension to be less than 36'' since the special poster printers use paper rolls of width 36''.
  • You can create a bunch of "normal" presentation slides, print out each one on a piece of (letter-sized) paper, and put them all together on a poster board. This is somewhat messier (and not as pretty) but you don't need to wait for the special poster printer to print.
• We will provide you with foam boards onto which you can tack on your poster (or your slides for the poster). The foam boards are 32'' by 40''.
• Printing: Unfortunately, we don't have a budget to pay for printing. If you are an SCS student, SCS has a poster printer you can use which prints on a 36" wide roll of paper. You have to call Operations and request to be added to the queue and they will call you back and tell you when you are able to print. It takes a while to print however especially if there are people ahead of you already printing so you can't wait until the last hour (or last day). But they are open 24/7. The printer sometimes runs out of paper, and you should really start early!!! The number of operations is 8-2608 or you can call the help desk (8-4231) to get the number.
  If you are a student outside SCS, you will need to check with your department to see if there are printing facilities for big posters (we're not sure what is offered outside SCS), or print a set of regular sized pages.

 

 

---

Datasets and project suggestions:   Below are descriptions of several data sets, and some suggested projects.  The first few are spelled out in greater detail.  You are encouraged to select and flesh out one of these projects, or make up you own well-specified project using these datasets.  If you have other data sets you would like to work on, we would consider that as well, provided you already have access to this data and a good idea of what to do with it.
 

Data Set Categories:

• fMRI Brain Imaging Data
• Image Data
• Sensor Network Data
• Text Data
• Datasets of Datasets
• Additional Data Sets

---

fMRI Brain Imaging Data

---

Brain imaging data (fMRI)

This data set contains a time series of images of brain activation, measured using fMRI, with one image every 500 msec. During this time, human subjects performed 40 trials of a sentence-picture comparison task (reading a sentence, observing a picture, and determining whether the sentence correctly described the picture). Each of the 40 trials lasts approximately 30 seconds. Each image contains approximately 5,000 voxels (3D pixels), across a large portion of the brain. Data is available for 12 different human subjects. 
Available software: we can provide Matlab software for reading the data, manipulating and visualizing it, and for training some types of classifiers (Gassian Naive Bayes, SVM).


Project A1: Bayes network classifiers for fMRI
Project idea: Gaussian Naïve Bayes classifiers and SVMs have been used with this data to predict when the subject was reading a sentence versus perceiving a picture. Both of these classify 8-second windows of data into these two classes, achieving around 85% classification accuracy [Mitchell et al, 2004]. This project will explore going beyond the Gaussian Naïve Bayes classifier (which assumes voxel activities are conditionally independent), by training a Bayes network in particular a TAN tree [Friedman, et al., 1997]. Issues youll need to confront include which features to include (5000 voxels times 8 seconds of images is a lot of features) for classifier input, whether to train brain-specific or brain-independent classifiers, and a number of issues about efficient computation with this fairly large data set. Midpoint milestone: By April 12 you should have run at least one classification algorithm on this data and measured its accuracy using a cross validation test. This will put you in a good position to explore refinements of the algorithm, alternative feature encodings for the data, or competing algorithms, by the end of the semester. Project: Reducing dimensionality and classification accuracy.
Papers to read: "Learning to Decode Cognitive States from Brain Images," Mitchell et al., 2004, "Bayesian Network Classifiers" Friedman et al., 1997.

Project A2: Dimensionality reduction for fMRI data
Project idea:  Explore the use of dimensionality-reduction methods to improve classification accuracy with this data.  Given the extremely high dimension of the input (5000 voxels times 8 images) to the classifier, it is sensible to explore methods for reducing this to a small number of dimension. For example, consider PCA, hidden layers of neural nets, or other relevant dimensionality reducing methods.  PCA is an example of a method that finds lower dimension representations that minimize error in reconstructing the data.  In contract, neural network hidden layes are lower dimensional representations of the inputs that minimize classification error (but only find a local minimum).  Does one of these work better?  Does it depend on parameters such as the number of training examples?
Papers to read: "Learning to Decode Cognitive States from Brain Images," Mitchell et al., 2004, papers and textbook on PCA, neural nets, or whatever you propose to try.

Project A3: Feature selection/feature invention for fMRI classification.
Project idea:  As in many high dimensional data sets, automatic selection of a subset of features can have a strong positive impact on classifier accuracy.  We have found that selecting features by the difference in their activity when the subject performs the task, relative to their activity while the subject is resting, is one useful strategy [Mitchell et al., 2004].  In this project you could suggest, implement, and test alternative feature selection strategies (eg., consider  the incremental value of adding a new feature to the current feature set, instead of scoring each feature independent of other features that are being selected), and see whether you can obtain higher classification accuracies.   Alternatively, you could consider methods for synthesizing new features (e.g., define the 'smoothed value' of a voxel in terms of a spatial Gaussian kernel function applied to it and its neighbors, or define features by averaging voxels whose time series are highly correlated).
Papers to read: "Learning to Decode Cognitive States from Brain Images," Mitchell et al., 2004, papers on feature selection

 

---

Image Data

---

Character recognition (digits) data

Optical character recognition, and the simpler digit recognition task, has been the focus of much ML research. We have three datasets on this topic. The first tackles the more general OCR task, on a small vocabulary of words: (Note that the first letter of each word was removed, since these were capital letters that would make the task harder for you.)

http://ai.stanford.edu/~btaskar/ocr/

The second dataset is the now "classic" digit recognition task for outgoing mail zip codes:

http://yann.lecun.com/exdb/mnist/

The third (and most challenging) data set consists of scrambled text known as Captchas that were designed by Luis Von Ahn to be difficult to automatically recognize. For more about Captchas go to Wikipedia Article or Captcha.net where you will find several papers.

Easier Dataset

Difficult Dataset

Project suggestions:

• Learn a classifier to recognize the letter/digit

• Use an HMM to exploit correlations between neighboring letters in the general OCR case to improve accuracy. (Since ZIP codes don't have such constraints between neighboring digits, HMMs will probably not help in the digit case.)

•  Apply a clustering/dimensionality reduction algorithm on this data, see if you get better classification on this lower dimensional space.

• Learn a classifier to decipher Captchas. You may want to begin by reading the following:
  • Using Machine Learning to Break Visual Human Interaction Proofs
  • Breaking a Visual CAPTCHA by Greg Mori and Jitendra Malik
  You may want to begin by building a classifier to segment the image into seperate letters.

---

Image Segmentation Dataset

The goal is to segment images in a meaningful way.  Berkeley collected three hundred images and paid students to hand-segment each one (usually each image has multiple hand-segmentations).   Two-hundred of these images are training images, and the remaining 100 are test images.  The dataset includes code for reading the images and ground-truth labels, computing the benchmark scores, and some other utility functions.  It also includes code for a segmentation example.  This dataset is new and the problem unsolved, so there is a chance that you could come up with the leading algorithm for your project.
http://www.cs.berkeley.edu/projects/vision/grouping/segbench/

Project ideas:
Project B1: Region-Based Segmentation
Most segmentation algorithms have focused on segmentation based on edges or based on discontinuity of color and texture.  The ground-truth in this dataset, however, allows supervised learning algorithms to segment the images based on statistics calculated over regions.  One way to do this is to "oversegment" the image into superpixels (Felzenszwalb 2004, code available) and merge the superpixels into larger segments.  Come up with a set of features to represent the superpixels (probably based on color and texture), a classifier/regression algorithm (suggestion: boosted decision trees) that allows you to estimate the likelihood that two superpixels are in the same segment, and an algorithm for segmentation based on those pairwise likelihoods. Since this project idea is fairly time-consuming focusing on a specific part of the project may also be acceptable. 
Milestone: By April 12, you should be able to estimate the likelihood that two superpixels are in the same segment and have a quantitative measure of how good your estimator is.  You should also have an outline of how to use the likelihood estimates to form the final segmentation.  The rest of the project will involve improving your likelihood estimation and your grouping algorithm, and in generating final results. 
Papers to read: Some segmentation papers from Berkeley are available here

Project B2: Supervised vs. Unsupervised Segmentation Methods
Write two segmentation algorithms (these may be simpler than the one above): a supervised method (such as logistic regression) and an unsupervised method (such as K-means). Compare the results of the two algorithms. For your write-up, describe the two classification methods that you plan to use.
Milestone: By April 12, you should have completed at least one of your segmentation algorithms and have results for that algorithm.  
Papers to read: Some segmentation papers from Berkeley are available here
 

---

Object Recognition

The Caltech 256 dataset contains images of 256 object categories taken at varying orientations, varying lighting conditions, and with different backgrounds.
http://www.vision.caltech.edu/Image_Datasets/Caltech256/

Project ideas:
 

• You can try to create an object recognition system which can identify which object category is the best match for a given test image.

• Apply clustering to learn object categories without supervision

Papers to read: See link.

 

---

Face recognition data 

There are two data sets for this problem. The first dataset contains 640 images of faces. The faces themselves are images of 20 former Machine Learning students and instructors, with about 32 images of each person. Images vary by the pose (direction the person is looking), expression (happy/sad), face jewelry (sun glasses or not), etc. This gives you a chance to consider a variety of classification problems ranging from person identification to sunglass detection. The data, documentation, and associated code are available here:

CMU Machine Learning Faces

Available Software: The same website provides an implementation of a neural network classifier for this image data. The code is quite robust, and pretty well documented in an associated homework assignment.

The second data set consists of 2253 female and 1745 male rectified frontal face images scraped from the hotornot.com website by Ryan White along with user ratings of attractiveness. The data set can be found here:

Facial Attractiveness Images.

Project ideas:
 

• Try SVM's on this data, and compare their performance to that of the provided neural networks

• Apply a clustering algorithm to find "similar" faces

• Learn a facial attractiveness classifier. A recent NIPS paper on the topic of predicting facial attractiveness can be found here.

---

Sensor Network Data

---

Sensor network data

 

 

Using this 54-node sensor network deployment, we collected temperature, humidity, and light data, along with the voltage level of the batteries at each node. The data was collected every 30 seconds, starting around 1am on February 28th 2004.

http://www-2.cs.cmu.edu/~guestrin/Research/Data/

This is a "real" dataset, with lots of missing data, noise, and failed sensors giving outlier values, especially when battery levels are low.

Project ideas:

• Compare regression algorithms

• Detect failed sensors

• Learn graphical models representing the correlations between measurements at different nodes

• Develop new distributed algorithms for solving a learning task on this data

Papers:

• http://www-2.cs.cmu.edu/~guestrin/Publications/IPSN2004/ipsn2004.pdf

• http://www-2.cs.cmu.edu/~guestrin/Publications/VLDB04/vldb04.pdf

Copyright (c) 2007 -- Carnegie Mellon University