Ricardo Silva is a Professor of Statistical Machine Learning and Data Science at University College London's Department of Statistical Science, specializing in computational causal inference and graphical models. His research bridges statistics, machine learning, and real-world applications across domains including algorithmic fairness, sports analytics, and network science. His primary research interests focus on causal inference methodologies , particularly counterfactual reasoning, latent variable models, and relational learning frameworks. Silva develops Bayesian computational approaches for causal discovery and effect estimation, with significant contributions to handling hidden confounders and unmeasured variables. His work frequently integrates graphical models with machine learning techniques to address complex inference problems in high-dimensional settings. Key research trends across his publications reveal a strong emphasis on Causal structure learning from observational and interventional data Counterfactual frameworks for algorithmic fairness Bayesian nonparametric methods for relational data Applications in sports analytics and network dynamics His methodological innovations often manifest as open-source software tools that have become standard in causal inference research. Ricardo actively contributes to the academic community through leadership roles in major conferences: Co-organized Uncertainty in Artificial Intelligence (UAI) 2018 and 2019 Chaired workshops at NIPS/NeurIPS on causal inference (2016-2017) Organized the ongoing Centre for Computational Statistics and Machine Learning Masterclasses at UCL As an educator and researcher, Silva maintains strong industry connections through software development and collaborative projects, particularly in applying causal methods to real-world decision systems. His teaching focuses on advanced statistical machine learning concepts with practical implementation components.
