Carlo Ciliberto is an Associate Professor in Machine Learning at University College London. His research focuses on theoretical and applied machine learning, with particular emphasis on structured prediction, meta-learning, optimal transport, and quantum computing. He has contributed to advancements in kernel methods, reinforcement learning, and robotics perception systems, notably through work with humanoid robots like the iCub. Key research interests include: Developing algorithms for distribution regression and Wasserstein-based learning. Exploring meta-learning frameworks for few-shot and incremental learning tasks. Designing robust systems for robotics applications, such as object recognition and tactile sensing. Investigating statistical foundations of quantum machine learning. Notable contributions to the field include the Manifold Structured Prediction framework, Sliced Wasserstein Kernels for distribution regression, and methodologies for conditional meta-learning. His work bridges theory and practice, with applications ranging from civil infrastructure analysis to humanoid robot perception.
