Efficient Second-Order Optimisation Algorithms for Learning from Big Data (2018-2023)

Scalable optimisation methods are now an integral part of machine learning (ML) in the presence of ¿big data¿. While the development of efficient first-order methods has seen explosive growth in the ML community, secondorder alternatives have been largely ignored. This is despite the fact that they are embraced by the scientific computing community. To bridge this gap, this project aims to apply a diverse range of techniques from scientific computing to design and implement new second-order methods that can surpass first-order alternatives in the next generation of optimisation methods for large-scale ML. The expected outcomes will facilitate the development of more effective ML algorithms for extraction of knowledge from large data sets.