I am a 3rd year DPhil Student under the supervision of Patrick Rebeschini and George Deligiannidis. Before Oxford, I was at "Sapienza" University of Rome, where I completed my BSc under the supervision of Enzo Orsingher.

My research focuses on theory for reinforcement learning. I am interested in building and analysing reinforcement learning algorithms using standard optimization tools, such as natural gradient descent and mirror descent. In particular, I am interested in settings where it is possible to take advantage of the structure of the environment, such as multi-agent settings or low-rank environments.

I am a 3rd year Phd student keen to better understand the influence of varying population size on population genetics models! Most models in population genetics assume the same population density and interactions at all space-time positions. In contrast, inhomogeneous dynamics are present in ecological models, which in turn often neglect the development of the genetic information carried by the population. With my current projects, I hope to make a step towards weakening this separation. Apart from mathematics, I enjoy running and memory competitions. Feel free to get in touch!

• Stochastic Processes
• Population Genetics

I am a DPhil student at the University of Oxford supervised by Patrick Rebeschini and Arnaud Doucet. Before this, I studied Mathematics at Imperial College London, supervised by Grigorios A. Pavliotis. I’m interested in stochastic optimisation, MCMC and theoretical foundations for machine learning. Most recently, I have been interested in the use of noise for regularisation and the relationship between optimisation and sampling.

Stochastic optimization, MCMC, generalization bounds, Langevin dynamics

I am a DPhil student at the department of statistics, University of Oxford. Currently, I am working on Stein's characterizations of distribution with particular emphasis on Bernoulli mixture graphs under the supervision of Professor Gesine Reinert. I am also a lecturer in statistics at the department of statistics, Lahore College for Women University, Lahore (2017 - Present; currently on study leave). I also worked as a lecturer in statistics at Queen Mary College, Lahore (2015 - 2017) and as a visiting teaching assistant at the department of statistics, Lahore College for Women University (2014 - 2015). In my MS in Statistics (2013) at Lahore College for Women University, Lahore I worked on Generalized Poisson-Exponential distribution; in my BS in Statistics (2011), I worked on Distributional properties of Generalized order statistics for Extended-Exponential distribution. I also worked on Extended Poisson exponential distribution and Transmuted exponentiated Pareto-I distribution.

• Probability Distributions
• Stein's Method
• Networks

I am a third year StatML student in the Department of Statistics at the University of Oxford.

My research interests include generative models, variational inference and MCMC. Recently, I have worked on denoising generative models for discrete data, viewing the process in continuous time and simulating with chemical physics inspired integrators. Previously, I have worked on online variational inference for sequential state space models by using links with reinforcement learning. I have also explored the links between variational inference and MCMC methods for more efficient sampling with applications to sampling molecular configurations from the Boltzmann distribution.

Causal inference, explainable AI, deep probabilistic learning, information theory, Bayesian statistics.

I’m a Florence Nightingale Bicentennial Fellow at the Department of Statistics, University of Oxford. Prior to that I was a graduate student on the StatML CDT programme at the University of Oxford, working with Tom Rainforth and Yee Whye Teh.

During my PhD I’ve interned as a Research Scientist at Microsoft Research Cambridge, where I focused on causal machine learning, and at Meta AI (FAIR Labs) NYC, where I worked on neural data compression. Before StatML, I spent four years in quant finance – first in quantitative equity research at UBS and later in cross-asset systematic trading strategies structuring at Goldman Sachs.

I'm broadly interested in probabilistic machine learning. I have worked on Bayesian experimental design, causality and neural data compression. Nowadays I'm mostly interested in robust evaluations of language models ("LLM Evals") and uncertainty quantification for LLMs.

I occasionally blot at Probably Approximately Incorrect.

I am a doctoral student in the Modern Statistics and Statistical Machine Learning (StatML) CDT, interested in developing statistical and machine learning methods for population genetics leveraging ancient DNA data. I am particularly motivated by healthcare-related applications of Machine Learning and genetics. I have also worked on developing deep learning algorithms for disease diagnosis from medical imaging scans. Prior to my DPhil in Oxford, I completed a Bachelor’s in Mathematics and an MPhil in Epidemiology & Biostatistics (focused on breast cancer genetic epidemiology), both at the University of Cambridge.

Statistical and population genetics, ancient DNA, genealogical inference and its applications to the detection of association and natural selection, machine learning for healthcare, biomedical imaging

I am a DPhil student on the CDT Mathematics of Random Systems, a program that is jointly run by the University of Oxford and Imperial College London. My research is focused on continuous-time branching processes, and my expected date of completion is October 2024. Before starting my DPhil, I obtained a MMath in Mathematics at the University of Oxford in 2020.

• Branching processes
• Lévy processes
• Interval Partitions
• Diffusions
• Continuum random trees
• Scaling limits

My research focuses on statistical methods to analyse how biomarkers affect the risk of disease (joint modelling of longitudinal and time-to-event data).

I am developing a joint model to understand how Covid-19 antibodies affects the risk of infection after receiving a vaccine.

I graduated from a Masters in Mathematics and Statistics from the University of Oxford in 2020. I then worked as a Statistician on the Covid-19 vaccine trials at the Oxford Vaccine Group, before starting my DPhil in 2021.

• Joint modelling of longitudinal and time-to-event data
• Survival analysis
• Bayesian modelling
• Multiple imputation

I am interested in developing flexible joint models using which can scale to large datasets. I am also keen to learn more about competing risks, interval censoring, causal inference and spatial statistics.

Please get in touch if you're interested in collaborating, or just want to chat!

Phillips, D. J. and Christodoulou, M. D. and Feng, S. and Pollard, A. J. and Voysey, M. and Steinsaltz, D. Improved estimates of COVID-19 correlates of protection, antibody decay and vaccine efficacy waning: a joint modelling approach. medRxiv (2024).