kangzi / OnlineLDS

Source code for the AAAI 2019 paper "On-Line Learning of Linear Dynamical Systems: Exponential Forgetting in Kalman Filters" (https://arxiv.org/abs/1809.05870)

OnlineLDS

Source code for the AAAI 2019 paper "On-Line Learning of Linear Dynamical Systems: Exponential Forgetting in Kalman Filters" (https://arxiv.org/abs/1809.05870). If you use this code, please cite the paper as:

@inproceedings{kozdoba2018,
 title={On-Line Learning of Linear Dynamical Systems: Exponential Forgetting in Kalman Filters},
  author={Kozdoba, Mark and Marecek, Jakub and Tchrakian, Tigran and Mannor, Shie},
  booktitle = {The Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19)},
  note={arXiv preprint arXiv:1809.05870},
  year={2019}
}

Running experiments.py recreates the plots used in the final 8-page version of the paper. Further calls explained therein produce further illuminating plots, which did not make it into the final 8-page version of the paper. Some of these require data from Prof. Steve Hoi in ./OARIMA_code_data/ which are distributed by an independent, permissive license.

To reuse the algorithms introduce by ourselves, see onlinelds.py for an implementation.