This summer, the Centre for Mathematical Imaging Techniques (CMIT) ran an intern programme for undergraduate students on "deep learning for mathematical imaging”. The internship provides opportunities for Year 1-3 students to engage in hands-on experience in research at the interface of mathematics, computing, and data science.
This year’s summer school brought together a record 28 participants for an engaging training program led by invited experts from the Universities of Liverpool, Strathclyde, Oxford, Bath, Lancaster, and Birmingham, along with a specialist from our industrial partner, MathWorks. During the research phase, students collaborated in small groups to apply deep learning methods to challenging problems in grain imaging (materials science), medical imaging, and tomography, exploring how mathematics, algorithms, and data interact in these applications.
We organized a series of online lectures on Python basics, deep learning with PyTorch, Google’s Colab, reinforcement learning, attention networks, generative adversarial networks (GANs), and machine learning basics with Matlab. Lectures were given by Dr. Andreas Alpers (UoL), Dr. Aigerim Saken (UoL), Prof. Ke Chen (University of Strathclyde), Dr. Tao Du (University of Oxford), Dr. Liam Burrows (University of Bath), Dr. Maciej Buze (Lancaster University), Dr. Luoying Hao (University of Birmingham) and Dr. George Amarantidis Koronaios (MathWorks). Congratulations to Zhiwen Cheng, Kunyu Duan, Eilidh Giffen, Shiyao Gu, Gina Gundersen, Yanjie Huang, Jierui Li, Yufan Lin, Haowen Lu, Jonathan Stam, and Jianxuan Zhou, who successfully completed their projects with distinction and received certificates of achievement.
Building on the positive experience and strong interest, we will offer the internship again next summer, giving students another opportunity to deepen their research experience in the rapidly evolving field of data science.