Awesome

For the latest update, please go to DeepCriticalLearning

Deep Robustness: ProSelfLC-CVPR 2021 + Example Weighting (DM+IMAE)

ProSelfLC: Progressive Self Label Correction for Training Robust Deep Neural Networks

• Paper link: https://arxiv.org/abs/2005.03788

Mean Absolute Error Does Not Treat Examples Equally and Gradient Magnitude’s Variance Matters

Derivative Manipulation: Example Weighting via Emphasis Density Funtion in the context of DL

• Novelty: moving from loss design to derivative design

For any specific discussion or potential future collaboration, please feel free to contact me. <br />

Please kindly cite the following papers if you find this repo useful.

@inproceddings{wang2021proselflc,
  title={ {ProSelfLC}: Progressive Self Label Correction
  for Training Robust Deep Neural Networks},
  author={Wang, Xinshao and Hua, Yang and Kodirov, Elyor and Clifton, David A and Robertson, Neil M},
  booktitle={CVPR},
  year={2021}
}
@phdthesis{wang2020example,
  title={Example weighting for deep representation learning},
  author={Wang, Xinshao},
  year={2020},
  school={Queen's University Belfast}
}
@article{wang2019derivative,
  title={Derivative Manipulation for General Example Weighting},
  author={Wang, Xinshao and Kodirov, Elyor and Hua, Yang and Robertson, Neil},
  journal={arXiv preprint arXiv:1905.11233},
  year={2019}
}
@article{wang2019imae,
  title={{IMAE} for Noise-Robust Learning: Mean Absolute Error Does Not Treat Examples Equally and Gradient Magnitude’s Variance Matters},
  author={Wang, Xinshao and Hua, Yang and Kodirov, Elyor and Robertson, Neil M},
  journal={arXiv preprint arXiv:1903.12141},
  year={2019}
}

PyTorch Implementation for ProSelfLC, Derivative Manipulation, Improved MAE

• Easy to install
• Easy to use
• Easy to extend: new losses, new networks, new dataset and loaders
• Easy to run experiments and sink results
• Easy to put sinked results into your technical reports and academic papers.

Install

Set the Pipenv From Scratch

• sudo apt update && sudo apt upgrade
• sudo apt install python3.8
• curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
• python3.8 get-pip.py
• vim ~/.bashrc -> add export PATH="/home/ubuntu/.local/bin:$PATH" -> source ~/.bashrc
• pip3 install pipenv

Build env for this repo using pipenv

• git clone git@github.com:XinshaoAmosWang/ProSelfLC-CVPR2021.git
• cd ProSelfLC-CVPR2021
• pipenv install -e .

How to use

Run experiments

• cd ProSelfLC-CVPR2021
• pipenv shell
• python tests/convnets/test_trainer_cnn_vision_cifar100_cce_convets.py
• python tests/convnets/test_trainer_cnn_vision_cifar100_lscplc_covnets.py
• python tests/convnets/test_trainer_cnn_vision_cifar100_proselflc_covnets.py
• python tests/derivativemanipulation_imae/test_dm_imae.py

Visualize results

• The results are well sinked and organised, e.g., proselflc_experiments/cifar100_symmetric_noise_rate_0.2/shufflenetv2

How to extend this repo

• Add dataset and dataloader: see examples in src/proselflc/slices/datain
• Add losses: see examples in src/proselflc/slices/losses
• Add networks: see examples in src/proselflc/slices/networks
• Add optimisers: see examples in src/proselflc/optim
• Extend the slicegetter: src/proselflc/slicegetter
• Write run scripts: see examples in tests/convnets/

Examples of sinked experimental configs and resutls

• The results are well sinked and organised, e.g., proselflc_experiments/cifar100_symmetric_noise_rate_0.2/shufflenetv2

• 

• 

• accuracy_loss.xlsx

• params.csv

Link to Slide, Poster, Final version

Link to reviewers' comments

List of Content

0. Storyline
1. Open ML Research Questions
2. Noticeable Findings
3. Literature Review
4. In Self LC, a core question is not well answered
5. Underlying Principle of ProSelfLC
6. Mathematical Details of ProSelfLC
7. Design Reasons of ProSelfLC
8. Related Interesting Work