Victor Khamesi

I’m a third-year PhD student in Statistics and Machine Learning at Imperial College London, supervised by Dean Bodenham, Ed Cohen, and Niall Adams. I’m also currently an Applied Science PhD Intern at Microsoft working with Jorge Tavares on generative language models for retrieval.

Prior to my PhD, I graduated with an MSc in Statistics at Imperial College London and an MSc in General Engineering (“Ingénieur Généraliste”) at École Centrale de Lyon. I previously worked as a Data Scientist at Amazon in Edinburgh, building machine learning models for behavioural and contextual targeted advertising.

Research interests. My work focuses on developing statistical and machine learning methods for detecting changes and anomalies in the generative process of sequential data. I am interested in online, unsupervised, and adaptive learning for high-dimensional and non-i.i.d. data, with methods that generalise across domains and modalities. My broader goal is to develop principled methods for improving learning algorithms in non-stationary environments, including continual learning, reinforcement learning, and sequential decision-making.