Masrour Zoghi

Papers

Total Citations

629

H-Index

About

Masrour Zoghi is a researcher whose work sits at the intersection of machine learning, optimization, and intelligent systems, with a particular focus on scaling Bayesian optimization to real-world, high-dimensional problems. His most influential contributions address one of the field's longstanding limitations: the inability of Bayesian optimization methods to perform efficiently beyond problems of moderate dimensionality. Through the innovative application of random embeddings, Zoghi and his collaborators demonstrated that Bayesian optimization could be extended to problems involving billions of dimensions — a breakthrough with profound implications for robotics, sensor placement, recommendation systems, advertising, and automatic algorithm configuration. His 2016 paper on this approach has accumulated 372 citations, underscoring its significance within the machine learning community, while his earlier 2013 work on the same theme has garnered an additional 242 citations — reflecting sustained and growing interest in his ideas over time. Together, these works represent a coherent and impactful research agenda that has helped unlock Bayesian optimization for practitioners working on large-scale, complex systems. For students exploring hyperparameter tuning, AutoML, or scalable probabilistic methods, Zoghi's research offers both theoretical insight and practical inspiration.