@misc{16461,
  author = {Rohan Money and Joshin Krishnan and Baltasar Beferull-Lozano},
  title = {Random Feature Approximation for Online Nonlinear Graph Topology Identification},
  abstract = {Online topology estimation of graph-connected time series is challenging, especially since the causal dependencies in many real-world networks are nonlinear. In this paper, we propose a kernel-based algorithm for graph topology estimation. The algorithm uses a Fourier-based Random feature approximation to tackle the curse of dimensionality associated with the kernel representations. Exploiting the fact that the real-world networks often exhibit sparse topologies, we propose a group lasso based optimization framework, which is solved using an iterative composite objective mirror descent method, yielding an online algorithm with fixed computational complexity per iteration. The experiments conducted on real and synthetic data show that the proposed method outperforms its competitors.},
  year = {2021},
  journal = {2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) 2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)},
  month = {10/2021},
  publisher = {IEEE},
  address = {Gold Coast, Australia},
  url = {https://ieeexplore.ieee.org/document/9596512},
  doi = {10.1109/MLSP52302.2021.9596512},
  note = {This work was carried out at University of Agder with the funding from the IKTPLUSS INDURB grant 270730/O70 and the SFI Offshore Mechatronics grant 237896/O30 from the Research Council of Norway.},
}