@inproceedings{discovery10166658,
          series = {International Conference on Neural Information Processing Systems},
           month = {December},
         address = {Vancouver, Canada},
          editor = {H. Larochelle and M. Ranzato and R. Hadsell and M.F. Balcan and H. Lin},
            year = {2020},
         journal = {Advances in Neural Information Processing Systems},
       publisher = {Neural Information Processing Systems Conference},
            note = {This version is the version of record. For information on re-use, please refer to the publisher's terms and conditions.},
          volume = {2020},
       booktitle = {NIPS'20: Proceedings of the 34th International Conference on Neural Information Processing Systems},
           title = {A Non-Asymptotic Analysis for
Stein Variational Gradient Descent},
            issn = {1049-5258},
        abstract = {We study the Stein Variational Gradient Descent (SVGD) algorithm, which optimises a set of particles to approximate a target probability distribution {\ensuremath{\pi}} ? e
?V
on R
d. In the population limit, SVGD performs gradient descent in the space
of probability distributions on the KL divergence with respect to {\ensuremath{\pi}}, where the
gradient is smoothed through a kernel integral operator. In this paper, we provide a
novel finite time analysis for the SVGD algorithm. We provide a descent lemma
establishing that the algorithm decreases the objective at each iteration, and rates
of convergence for the averaged Stein Fisher divergence (also referred to as Kernel
Stein Discrepancy). We also provide a convergence result of the finite particle
system corresponding to the practical implementation of SVGD to its population
version.},
             url = {https://proceedings.neurips.cc/paper/2020/hash/3202111cf90e7c816a472aaceb72b0df-Abstract.html},
          author = {Korba, A and Salim, A and Arbel, M and Luise, G and Gretton, A}
}