I am a Research Fellow working with Prof. Wilfrid Kendall. We are exploring the use of Dirichet forms to address optimal scaling and other theoretical questions about Markov chain Monte Carlo methods. From April 2017 until September 2018 I was a postdoctoral research assistant at Queen Mary University of London with Prof. John Moriarty, working on simulation of rare events and other applications of MCMC methods to energy related problems. I did my PhD (2013-2017) at Imperial College London with Prof. Aleksandar Mijatović on the topic of variance reduction using control variates obtained by approximately solving the Poisson equation.
Some Publications and Preprints:
- Vogrinc J, and Kendall WS. "Counterexamples for optimal scaling of Metropolis-Hastings chains with rough target densities." (2019). arXiv
- Moriarty J, Vogrinc J, and Zocca A. "The Skipping Sampler: A new approach to sample from complex conditional densities." (2019). arXiv.
- Mijatović A, and Vogrinc J. "Asymptotic variance for Random Walk Metropolis chains in high dimensions: logarithmic growth via the Poisson equation." To appear in Advances in Applied Probability 51.4 (December 2019). arXiv.
- Moriarty J, Vogrinc J, and Zocca A. "Frequency violations from random disturbances: an MCMC approach." In 2018 IEEE Conference on Decision and Control (CDC), pp. 1598-1603. IEEE, 2018. arXiv.
- Mijatović A, and Vogrinc J. "On the Poisson equation for Metropolis-Hastings chains." Bernoulli 24.3 (2018): 2401-2428. arXiv.
Mathematical Sciences Building
Rm MB2.13, Dept of Statistics
University of Warwick
Coventry, CV4 7AL