{"id":31,"date":"2019-11-29T15:49:21","date_gmt":"2019-11-29T15:49:21","guid":{"rendered":"http:\/\/www.lancaster.ac.uk\/stor-i-student-sites\/tamas-papp\/?page_id=31"},"modified":"2025-05-31T09:58:43","modified_gmt":"2025-05-31T09:58:43","slug":"research","status":"publish","type":"page","link":"https:\/\/www.lancaster.ac.uk\/stor-i-student-sites\/tamas-papp\/research\/","title":{"rendered":"Research"},"content":{"rendered":"\n

Publications<\/h2>\n\n\n\n
    \n
  1. Tam\u00e1s P. Papp and Chris Sherlock, Scalable couplings for the random walk Metropolis algorithm<\/em>, JRSS:B<\/strong>, 2024+\u00a0(journal<\/a>, arXiv<\/a>, code<\/a>)<\/li>\n\n\n\n
  2. Tam\u00e1s P. Papp, Paul Fearnhead and Chris Sherlock, Comments on “Denoising Markov Models” by J. Benton et al<\/em>., JRSS:B<\/strong>, 2024\u00a0(<\/a>journal<\/a>)<\/li>\n\n\n\n
  3. Tam\u00e1s P. Papp and Chris Sherlock, Centered plug-in estimation of Wasserstein distances<\/em>\u00a0(arXiv<\/a>, code<\/a>)<\/li>\n<\/ol>\n\n\n\n

    PhD<\/h2>\n\n\n\n

    Projects I have worked on include the design of efficient implementable couplings for Markov chain Monte Carlo (MCMC) algorithms, sensible parameter tuning for unbiased MCMC in the framework of Jacob et al (2020), and a debiasing procedure for empirical Wasserstein distance estimates. My PhD supervisor is Chris Sherlock.<\/p>\n\n\n\n

    MRes<\/h2>\n\n\n\n

    I wrote my dissertation on methodology and theory for unbiased Markov chain Monte Carlo, supervised by Chris Sherlock. Much of my PhD research stems from preliminary work carried out there.<\/p>\n\n\n\n

    Three literature reviews I wrote during the MRes are available below:<\/p>\n\n\n\n

      \n
    1. Bayesian nonparametrics<\/a> – supervised by Marco Battiston<\/a><\/li>\n\n\n\n
    2. Rate function estimation for non-homogeneous Poisson processes<\/a> – supervised by Lucy Morgan<\/a><\/li>\n\n\n\n
    3. Multi-armed bandits and Bayesian optimisation<\/a> – supervised by David S. Leslie<\/a><\/li>\n<\/ol>\n\n\n\n

      References<\/strong><\/h3>\n\n\n\n