Home

Awesome

[KDD 2023] Source-Free Domain Adaptation with Temporal Imputation for Time Series Data [Paper] [Cite]

by: Mohamed Ragab, Emadeldeen Eldele, Min Wu, Chuan-Sheng Foo, Xiaoli Li, Zhenghua Chen <br/>

Accepted in the 29th SIGKDD Conference on Knowledge Discovery and Data Mining - Research Track.

<p align="center"> <img src="misc/temporal_adapt.PNG" width="900" class="center"> </p>

Requirmenets:

Datasets

Available Datasets

We used four public datasets in this study. We also provide the preprocessed versions as follows:

Training procedure

The experiments are organised in a hierarchical way such that:

Training a model

To train a model:

python trainers/train.py  --experiment_description exp1  \
                --run_description run_1 \
                --da_method MAPU \
                --dataset HAR \
                --backbone CNN \
                --num_runs 3 \

Launching a sweep

Sweeps here are deployed on Wandb, which makes it easier for visualization, following the training progress, organizing sweeps, and collecting results.

python trainers/sweep.py  --experiment_description exp1_sweep  \
                --run_description sweep_over_lr \
                --da_method MAPU \
                --dataset HAR \
                --backbone CNN \
                --num_runs 3\
                --num_sweeps 50 \
                --sweep_project_wandb MAPU_HAR

Upon the run, you will find the running progress in the specified project page in wandb.

Results

Citation

If you found this work useful for you, please consider citing it.

@inproceedings{mpau,
  author = {Ragab, Mohamed and Eldele, Emadeldeen and Foo, Chuan-Sheng and Wu, Min and Li, Xiaoli and Chen, Zhenghua},
  title = {Source-Free Domain Adaptation with Temporal Imputation for Time Series Data},
  booktitle={29th SIGKDD Conference on Knowledge Discovery and Data Mining - Research Track},
  year = {2023},
  url={https://openreview.net/forum?id=v6GK0ijPW0B}
}