Michelangelo Conserva

profile picture 

"Caminante, no hay camino: se hace camino al andar."

Ph.D. candidate at Queen Mary University of London, United Kingdom.

Bayesian statistician off on an adventure in the land of Reinforcement Learning.

Email: m.conserva@qmul.ac.uk

Links:

Professional Experience

Spatially-aware graph neural networks for multivariate time series forecasting with a learned graph structure
Research intern at Google Research, Kenya
Supervisor: Aisha Walcott Bryant
June 2023 - Oct 2023

Publications

Posterior Sampling for Deep Reinforcement Learning
R. Sasso, M. Conserva, and P. Rauber
ICML 2023

Markov Decision Processes Hardness: Theory and Practice
M. Conserva and P. Rauber
NeurIPS 2022

Recurrent Neural-Linear Posterior Sampling for Non-Stationary Contextual Bandits
A. Ramesh, M. Conserva, P. Rauber, and J. Schmidhuber
Neural Computation 2022

The Graph Cut Kernel for Ranked Data
M. Conserva, M. Deisenroth, and K. S. S. Kumar
Transactions on Machine Learning Research 2022

Education

  • Ph.D. in Computer Science, Queen Mary University of London, UK, 2020 - 2024.

  • MSc in Computational Statistics and Machine Learning, University College London, UK, 2019 - 2020, Distinction.

  • BSc in Statistic, Economics and Finance, Sapienza University of Rome, 2016 - 2019, 110 cum laude.

Pictures

alt text