jtt-m

This is an implementation of our paper: Outlier-Aware Training for Improving Group Accuracy Disparities. If you find our code useful, please cite:

@inproceedings{chen-etal-2022-outlier,
    title = "Outlier-Aware Training for Improving Group Accuracy Disparities",
    author = "Chen, Li-Kuang  and
      Kruengkrai, Canasai  and
      Yamagishi, Junichi",
    booktitle = "Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing: Student Research Workshop",
    year = "2022",
    pages = "54--60",
    url = "https://aclanthology.org/2022.aacl-srw.8",
}

Requirements

The code is tested on Python 3.9 and PyTorch 1.10.1. We recommend to create a new environment for experiments using conda:

conda create -y -n jtt-m python=3.9
conda activate jtt-m

Then, from the jtt-m project root, run:

pip install -r requirements.txt

For further development or modification, we recommend installing pre-commit:

pre-commit install

To ensure that PyTorch is installed and CUDA works properly, run:

python -c "import torch; print(torch.__version__); print(torch.cuda.is_available())"

We should see:

1.10.1+cu111
True

⚠️ We use PyTorch 1.10.1 with CUDA 11.0. You may need another CUDA version suitable for your environment.

Experiments

See experiments.

Acknowledgments

This work is supported by JST CREST Grants (JPMJCR18A6 and JPMJCR20D3) and MEXT KAKENHI Grants (21H04906), Japan.

Licence

BSD 3-Clause License

Copyright (c) 2022, Yamagishi Laboratory, National Institute of Informatics All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

• Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

• Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

• Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.