FairFor

Source code for the paper, "Learning Informative Representation for Fairness-aware Multivariate Time-series Forecasting: A Group-based Perspective", accepted by TKDE.

Overview

In this work, we formulate the MTS fairness modeling problem as learning informative representations attending to both advantaged and disadvantaged variables. Accordingly, we propose a novel framework, named FairFor, for fairness-aware MTS forecasting, i.e., fair MTS forecasting.

Requirements

• Python 3.6
• matplotlib == 3.3.4
• numpy == 1.19.5
• pandas == 1.1.5
• scikit_learn == 0.24.2
• torch == 1.8.0

Datasets

• PeMSD7(M) - https://dot.ca.gov/programs/traffic-operations/mpr/pems-source
• Solar-Energy - http://www.nrel.gov/grid/solar-power-data.html
• Traffic - https://archive.ics.uci.edu/ml/datasets/PEMS-SF
• ECG5000 - http://www.timeseriesclassification.com/description.php?Dataset=ECG5000

Baselines

• LSTNet - https://github.com/laiguokun/LSTNet
• TPA-LSTM - https://github.com/shunyaoshih/TPA-LSTM
• TS2VEC - https://github.com/yuezhihan/ts2vec
• Informer - https://github.com/zhouhaoyi/Informer2020
• Pyraformer - https://github.com/ant-research/Pyraformer
• MTGNN - https://github.com/nnzhan/MTGNN
• StemGNN - https://github.com/microsoft/StemGNN
• AGCRN - https://github.com/LeiBAI/AGCRN

Citation

If you find our work useful, please consider citing the following paper

@article{DBLP:journals/corr/abs-2301-11535,
  author       = {Hui He and
                  Qi Zhang and
                  Shoujin Wang and
                  Kun Yi and
                  Zhendong Niu and
                  Longbing Cao},
  title        = {Learning Informative Representation for Fairness-aware Multivariate
                  Time-series Forecasting: {A} Group-based Perspective},
  journal      = {CoRR},
  volume       = {abs/2301.11535},
  year         = {2023}
}