Boosting Neural Networks

This goal of this project is to compare the capabilities of a cascade of shallow neural networks relative to a single deep neural network.
Type of Data: 
CIFAR100 images
Domain Expert: 
Methods Expert: 
Methods Student: 
Methods Student Openings: 
2.00
Methods Student Funding: 
no
Methods Student Prerequisites: 
Student should be experienced with tensorflow, and understand the basic theory of boosting as described in the papers: * Improved Boosting Algorithms Using Confidence-rated Predictions: https://link.springer.com/article/10.1023/A:1007614523901 * Boosting the margins: https://projecteuclid.org/journals/annals-of-statistics/volume-26/issue-5/Boosting-the-margin--a-new-explanation-for-the-effectiveness/10.1214/aos/1024691352.full