![The Queen's Tower](https://pxl-imperialacuk.terminalfour.net/fit-in/1079x305/prod01/channel_2/media/migration/giving/imp-130820-campus1-lib-001-136474-001_1615546110081_x4.jpg)
Thank you for supporting the Alvin Li-Shen Chua memorial scholarship
The scholarship is in memory of Alvin. The first awards were made in September 2022 to two postgraduate students on the MRES in Machine Learning and Big Data in the Physical Sciences, and further awards are made each September. The course reflects Alvin's academic and real-world scientific interests. Each student will receive £7,500 towards their tuition fees.
Further information can be found on Alvin’s scholarship website.
Make your gift
Alvin was an undergraduate of the Physics Department from 1997-2000 and during this time he completed a BSc theoretical project under the supervision of Prof. Kim Christensen, resulting in a research paper, a rarity for UG students. After graduating with a 1st class BSc degree in 2000, Alvin then undertook a PhD in the Condensed Matter Theory group under the supervision of Prof. Dimitri Vvedensky. He graduated in 2004 with a PhD thesis entitled "Mathematical Aspects of Epitaxial Systems".
He later worked as a Research Assistant with Dimitri on the mathematics of surface growth models (2006-2007) and then with Prof. Adrian Sutton and Prof. Mike Finnis on simulations of materials (2007-2009) leading to an important publication in the top journal Nature Materials in 2010. Alvin was a wonderful personality with a huge wide smile and a sharp scientific and analytic mind, allowing him to produce research results in his own unique way. Alvin is remembered fondly by his fellow students and colleagues here at Imperial.
Alvin’s interests
The fields of artificial intelligence and machine learning have seen tremendous progress in the past few years, followed by an exponential increase in the investments and creation of new jobs. Today we have conversations with AI bots, our phones can recognise objects and people in our photos, and our cars drive by themselves. This progress has been driven mostly by the growth in data, computing resources and, to a smaller extent, new theoretical ideas. A large number of machine learning algorithms show excellent performance, but they are neither very robust nor predictable. More theoretical work is necessary to understand those algorithms, improve them and allow further progress.
Machine learning researchers have usually a background in information engineering, computer science and statistics/math. The contribution of theoretical physicists is still negligible, but we think that a field growing at such a fast pace would benefit from interdisciplinary work. This effort is necessary not only to stimulate research at the academic level, but also to meet the increasing demand of companies for diverse expertise. We believe that theoretical physicists have a lot to offer to machine learning given their knowledge of advanced applied techniques such as statistical mechanics, quantum mechanics, dynamical systems, stochastic processes. They have also a lot to gain, in terms of new jobs and novel problems to study.