Accelerated Iterated Filtering
Simulation-based inferences have attracted much attention in recent years, as the direct computation of the likelihood function in many real-world problems is difficult or even impossible. Iterated filtering (Ionides, Bretó, and King 2006; Ionides, Bhadra, Atchadé,
and King 2011) enables maximization of likelihood function via model perturbations and approximation of the gradient of loglikelihood through sequential Monte Carlo filtering. By an application of Stein’s identity, Doucet, Jacob, and Rubenthaler (2013) developed a
second-order approximation of the gradient of log-likelihood using sequential Monte Carlo smoothing. Based on these gradient approximations, we develop a new algorithm for maximizing the likelihood using the Nesterov accelerated gradient. We adopt the accelerated inexact gradient algorithm (Ghadimi and Lan 2016) to iterated filtering framework, relaxing the unbiased gradient approximation condition. We devise a perturbation policy for iterated filtering, allowing the new algorithm to converge at an optimal rate for both concave and non-concave log-likelihood functions. It is comparable to the recently developed Bayes map iterated filtering approach and outperforms the original iterated filtering approach.
How to Cite
The Austrian Journal of Statistics publish open access articles under the terms of the Creative Commons Attribution (CC BY) License.
The Creative Commons Attribution License (CC-BY) allows users to copy, distribute and transmit an article, adapt the article and make commercial use of the article. The CC BY license permits commercial and non-commercial re-use of an open access article, as long as the author is properly attributed.
Copyright on any research article published by the Austrian Journal of Statistics is retained by the author(s). Authors grant the Austrian Journal of Statistics a license to publish the article and identify itself as the original publisher. Authors also grant any third party the right to use the article freely as long as its original authors, citation details and publisher are identified.
Manuscripts should be unpublished and not be under consideration for publication elsewhere. By submitting an article, the author(s) certify that the article is their original work, that they have the right to submit the article for publication, and that they can grant the above license.