Virginia Aglietti
News
- I am currenlty part of the Microsoft Research Lab in Cambridge for a summer internship (07/20-12/20) under the supervision of Javier Gonzalez.
I am currently a fourth year PhD student in Statistics supervised by Dr. Theodoros Damoulas. I am part of a joint PhD program between the Department of Statistics at the University of Warwick and the Department of Statistics at the University of Oxford called the Oxford - Warwick Statistics Programme (OxWaSP). I am part of the Warwick Machine Learning Group. I am currently collaborating with Dr. Edwin V. Bonilla on Multi-task learning of inhomogeneous point processes. More information about my background can be found on my CV.
Research Interests
My research interests are at the interface of Bayesian statistics and computer science. I am interested in developing scalable machine learning algorithms for multi-task learning with Gaussian processes. I am also interested in linking probabilistic models and real world decision making problems with the aim of developing probabilistic frameworks for analysing and assessing the impact of policies. In terms of applications, I am currently looking at spatio-temporal problems in social sciences, with a particular focus on crime.
Projects
- I completed a summer internship at Amazon Supply Chain Optimization Technologies (SCOT) group in Cambridge under the supervision of Javier Gonzalez.
- I am collaborating with the University of Sydney's Centre for Translational Data Science on Bayesian Optimisation for criminology. This is part of a broader collaboration between the Alan Turing Institute and the Centre for Translational Data Science.
- I visited CSIRO's Data61 (11/18-02/19) to work on Gaussian Process modulated Poisson Processes.
- I am a visiting researcher at the Alan Turing Institute and I am part of the Clean Air project. The project aims at developing statistical methodology and machine learning algorithms to support London's Major's office in taking data-driven/evidence based policy decisions in order to improve air quality over the city of London.
- I collaborated with CUSP London on a project aiming at assessing the impact of traffic policies on the number of accidents happening in London.
Publications
- Aglietti V., Damoulas T., Alvarez A.M., & Gonzalez J. (2020). Multi-task Causal Learning with Gaussian Processes,
Neural Information Processing Systems (NeurIPS) 2020. - Aglietti V., Lu X., Paleyes A., & Gonzalez J. (2020). Causal Bayesian Optimization, International Conference on Artificial Intelligence and Statistics (AISTATS) 2020.
- Aglietti V., Bonilla E., Damoulas T. & Cripps S. (2019). Structured Variational Inference in Continuous Cox Process Models, Neural Information Processing Systems (NeurIPS) 2019.
- Aglietti V., Damoulas T. & Bonilla E. (2019). Efficient Inference in Multi-task Cox Process Models, International Conference on Artificial Intelligence and Statistics (AISTATS) 2019.
- Calderoni F., Dugato M., Aglietti V., Aziani A. & Rotondi M. (2017). Price and Non-price Determinants of the Illicit Cigarette Trade: Analysis at the Subnational Level in the EU in Dual Markets: Comparative Approaches to Regulation.
Selected Talks & Posters
October 2018 (Talk): The Bridges seminar series, University of Warwick (Coventry, UK).
March 2018 (Poster): Amazon Machine Learning Workshop, Amazon Research (Berlin, Germany).
November 2018 (Talk): Variational Inference in Non-homogeneous Poisson Processes, Centre for Translational Data Science - The University of Sydney (Sydney, Australia).
April 2019 (Poster): Efficient Inference in Multi-task Cox Process Models, International Conference on Artificial Intelligence and Statistics (AISTATS) (Naha, Okinawa, Japan).
November 2019 (Talk): Amazon Machine Learning Workshop, Amazon Research (Cambridge,UK).
December 2019 (Poster): Structured Variational Inference in Continuous Cox Process Models, Neural Information Processing Systems (NeurIPS) (Vancouver, Canada)
June 2020 (Talk): Causal Decision Making Under Uncertainty, CogX (London, UK) (video)
September 2020 (Talk): Causal decision-making meets Gaussian Processes, Gaussian Processes Summer school (slides, video)
September 2020 (Talk): Multi-task Gaussian process modelling for point process data, Gaussian Processes Summer School (slides, video)
Software
The Python code for the published papers will be made available on my GitHub profile.
BSc/MSc Supervision
Every year I am supervising undergraduate and master's thesis projects at both Statistics and Computer Science departments.
Notable 1st Class MSc thesis:
- Edoardo Barp, MSc in Mathematics for Real World Systems, Bayesian Inverse Reinforcement Learning for Path-Based Reward Inference, Dept. of Mathematics, University of Warwick, 2018. [co-supervised with Theo Damoulas]
Teaching
- V.Aglietti at warwick.ac.uk
- aglietti at stats.ox.ac.uk
Location
Department of Statistics
University of Warwick
CV4 7AL, UK