This codebase compares two learning strategies Interleaving learning and Small Group Learning. Further, we also explore the effects of Small Group Learning on the COVID-CT dataset and improvement in image classification results. In order to run this code, please follow folders corresponding to the task of interest. Each folder contains a README.md to walk through the steps for running the code.
IL-darts- Trains a DARTS based Interleaving Learning model on CIFAR10/100SGL-pc-darts- Trains a PC-DARTS based Small Group Learning model on CIFAR10/100SGL-covid- Trains a PC-DARTS based Small Group Learning model on COVID-CT data
- Partial Channel Connections for Memory-Efficient Differentiable Architecture Search(PC-DARTS) by Yuhui Xu, Lingxi Xie, Xiaopeng Zhang, Xin Chen, Guo-Jun Qi, Qi Tian and Hongkai Xiong. Code, Paper
- Small Group Learning with Application to Neural Architecture Search by Xuefeng Du, Pengtao Xie. (Code provided privately) Paper
- COVID-CT-Dataset: A CT Scan Dataset about COVID-19 by Xingyi Yang, Xuehai He, Jinyu Zhao, Yichen Zhang, Shanghang Zhang, Pengtao Xie. Dataset, Paper
Work completed in partial fulfillment of course requirements for ECE 285, Deep Generative Models during Winter 2021 at UCSD by Aparna Srinivasan and Shreyas Rajesh