DAL with disentangled representations
PyTorch implementation for Domain agnostic learning with disentangled representations (ICML2019 Long Oral). This repository contains some code from Maximum Classifier Discrepancy for Domain Adaptation. If you find this repository useful for you, please also consider cite the MCD paper!
The code has been tested on Python 3.6+PyTorch 0.3 and Python 3.6+PyTorch 0.41. To run the training and testing code, use the following script:
python main.py --source=mnist --recordfolder=agnostic_disentangle --gpu=0
The poster of this paper can be found with the link: poster
The Oral presentation of this paper in ICML2019 can be found with the link: Oral Presentation
Dataset Download
Since many researchers have sent us emails for Digit-Five data. We share the Digit-Five dataset we use in our experiments in the following download link:
https://drive.google.com/open?id=1A4RJOFj4BJkmliiEL7g9WzNIDUHLxfmm
Keep in mind that the Mnist-M dataset is generated by ourselves, thus this subset may be different from the one released by DANN paper.
If you find the Digit-Five dataset useful for your research, please cite our paper.
Citation
If you use this code for your research, please cite our paper:
@inproceedings{Peng2019DomainAL,
title={Domain Agnostic Learning with Disentangled Representations},
author={Xingchao Peng and Zijun Huang and Ximeng Sun and Kate Saenko},
booktitle={ICML},
year={2019}
}