README.md

(IJCAI 2024) Bridging the Gap: Learning Pace Synchronization for Open-World Semi-Supervised Learning [Paper])

This repository contains the implementation details of our Learning Pace Synchronization (LPS) approach for Open-World Semi-Supervised Learning

Bo Ye, Kai Gan, Tong Wei, Min-Ling Zhang, "Bridging the Gap: Learning Pace Synchronization for Open-World Semi-Supervised Learning"\

If you use the codes from this repo, please cite our work. Thanks!

@inproceedings{ye2024bridging,
  author={Bo Ye and Kai Gan and Tong Wei and Min-Ling Zhang},
  title={Bridging the Gap: Learning Pace Synchronization for Open-World Semi-Supervised Learning},
  booktitle={Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, {IJCAI} 2024, Jeju, South Korea, August 3-9, 2024},
  pages        = {5362--5370},
  publisher    = {ijcai.org},
  year         = {2024}
}

Dependencies

The code is built with following libraries:

• PyTorch==1.9
• sklearn==1.0.1

Usage

Get Started

For ImageNet 100, you need to utilize 'gen_imagenet_list.py' to generate the corresponding sample's list.

And the pretraining weights used in our paper can be downloaded in this link, which is provided by ORCA.

• To train on CIFAR-10, run

python lps_cifar.py --dataset cifar10 --labeled-num 5 --labeled-ratio 0.5

• To train on CIFAR-100, run

python lps_cifar.py --dataset cifar100 --labeled-num 50 --labeled-ratio 0.5

• To train on ImageNet-100, run

python lps_imagenet.py --dataset imagenet100 --labeled-num 50 --labeled-ratio 0.5