TACTiS-2: Faster, Simpler Attentional Copulas for Multivariate Time Series

Arjun Ashok, Danil Korshunov, Karol Kole [TACTiS-2: Faster, Simpler Attentional Copulas for Multivariate Time Series]

We introduce a new model for multivariate probabilistic time series prediction, designed to flexibly address a range of tasks including forecasting, interpolation, and their combinations. Building on copula theory, we propose a simplified objective for the recently-introduced transformer-based attentional copulas (TACTiS), wherein the number of distributional parameters now scales linearly with the number of variables instead of factorially. The new objective requires the introduction of a training curriculum, which goes hand-in-hand with necessary changes to the original architecture. We show that the resulting model has significantly better training dynamics and achieves state-of-the-art performance across diverse real-world forecasting tasks, while maintaining the flexibility of prior work, such as seamless handling of unaligned and unevenly-sampled time series.

Installation

You can install the TACTiS-2 model with pip:

pip install tactis

Alternatively, the research version installs gluonts and pytorchts as dependencies which are required to replicate experiments from the paper:

pip install tactis[research]

Note: tactis has been currently tested with Python 3.10.8.

Instructions

With the research version of the code, train.py can be used to train the TACTiS-2 model for a specific dataset. The arguments in train.py can be used to specify the dataset, the training task (forecasting or interpolation), the hyperparameters of the model and a whole range of other training options.

There are notebooks in the that are useful in guiding training and evaluation pipeline setups: random_walk.ipynb demonstrates TACTiS-2 on a simple low-dimensional random walk dataset, and gluon_fred_md_forecasting.ipynb demonstrates how to train and evaluate TACTiS-2 on the FRED-MD dataset used in the paper. Note that the gluon_fred_md_forecasting.ipynb notebook requires GluonTS and PyTorchTS to be installed.

Note

For an implementation of the original version of TACTiS, please see here.

Citing this work

Please use the following Bibtex entry to cite TACTiS-2.

@misc{cody2023tactis2,
      title={TACTiS-2: Faster, Simpler Attentional Copulas for Multivariate Time Series}, 
      author={Arjun Ashok, Danil Korshunov, Karol Kole},
      year={2023},
      eprint={2310.01327},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}