Focused on supervised multi-modal machine learning, particularly the fusion of tabular data and high-dimensional data (such as natural text and images) to generate predictions with uncertainty bounds.
Education
PhD student (Business Analytics), University of Lausanne
Background
Ilia is a PhD student (business analytics) and lecturer (data science) at the University of Lausanne. His research focuses on supervised multi-modal machine learning, with tabular data as the primary modality and high-dimensional data (e.g., natural text and images) as secondary modalities. He studies the best ways of fusing different data sources to produce point predictions with uncertainty bounds (prediction intervals). His other research interests include latent representation (e.g., encoding, EM algorithm) and inferring with missing modalities.
Miscellany
Outside of his research, he pursues a wide range of activities for both physical and mental well-being. He has a keen passion for sports science and nutrition. He enjoys gaming (pcmasterrace) and building custom (GPU-poor) rigs. Occasionally, he invests in tech stocks, particularly NASDAQ and AI companies.