- Research Fellow (11/20-Present) — Queensland University of Technology (QUT), Centre for Data Science
- ACEMS Research Fellow (01/19 -10/ 20) — UNSW Sydney
- ACEMS Research Fellow (08/18 – 01 /19) [Short Contract] — The University of Queensland
- PhD Candidate in Statistics (2015-2018) — The University of Queensland.
- Advisor: Professor Dirk Kroese | Thesis: Advances in Monte Carlo Methodology
- 30/11/2021 – Presented at QUT Centre for Data Science “Data Science Under the Hood” series (video below).
If you enjoyed this talk, be sure to check out my course materials for my AMSI Winter School 2021 course on Deep Probabalistic Models (the above talk is a simplified version of the first quarter!).
My research, generally speaking, lies at the intersection of statistics and machine learning.
To date this has included topics such as inference methods for big data (especially time series), kernel methods, Markov Chain Monte Carlo sampling, variational inference, and rare-event simulation and probability estimation.
Some of my more recent interests moving forward include:
– Deep Probabilistic Programming/Modelling: (in particular with the Pyro language), that is, combining graphical models with deep learning and modern inference techniques.
– Deep Generative Models (Flow-Based Models, Diffusion models, Variational Autoencoders, and their generalizations such as Neural Processes).
– Representation Learning: How does one represent data in a simplistic way to capture its “essence” and facilitate learning (this includes non-linear approaches to dimensionality reduction and supervised/unsupervised disentanglement).
– The application of, and development of extensions to the above for solving complicated and interesting problems!
Sutton, M. , Salomone, R., Chevallier, A., and Fearnhead, P. (2022), Continuously-Tempered PDMP Samplers. arXiv:2205.09559
Villani, M., Quiroz, M., Kohn, R., and Salomone, R. (2021), Spectral Subsampling MCMC for Stationary Multivariate Time Series. arXiv:2104.02134
Hodgkinson, L., Salomone, R., and Roosta, F. (2021), The reproducing Stein kernel approach for post-hoc corrected sampling. arXiv: 2001.09266
Salomone, R., South, L.F., Drovandi, C.C., and Kroese, D.P. (2018), Unbiased and Consistent Nested Sampling via Sequential Monte Carlo. arXiv:1805.03924
Hodgkinson, L., Salomone,R., and Roosta, F. (2021), Implicit Langevin Algorithms for Sampling From Log-concave Densities, Journal of Machine Learning Research (JMLR) 22: 1-30. [Read Online]
Salomone R., Quiroz, M., Kohn, R., Villani, M., and Tran, M.N. (2020), Spectral Subsampling MCMC for Stationary Time Series, Proceedings of the International Conference on Machine Learning (ICML) 2020. [Read Online]
Botev, Z.I., Salomone, R., Mackinlay, D. (2019), Fast and accurate computation of the distribution of sums of dependent log-normals, Annals of Operations Research 280 (1), 19-46. [Read Online]
Laub, P.J., Salomone, R., Botev, Z.I. (2019), Monte Carlo estimation of the density of the sum of dependent random variables, Mathematics and Computers in Simulation 161, 23-31.
Salomone, R., Vaisman, R., and Kroese, D.P. (2016). Estimating the Number of Vertices in Convex Polytopes. Proceedings of the Annual International Conference on Operations Research and Statistics, ORS 2016. [Read Online]