skip to content

Faculty of Mathematics

 

I am a PhD student under supervision of Professor Carola Schönlieb at the Cambridge Image Analysis Group within DAMTP. I am also a member of the Cantab Capital Institute for the Mathematics of Information and the Maths4DL programme.

I am interested in research in a wide variety of areas of applied mathematics, ranging from inverse problems to machine learning, optimisation, theoretical inflationary cosmology and others. My current research focuses on the areas of geometric deep learning and physics informed machine learning.

I am currently looking for summer students in the areas described above and if any of those topics sound interesting - reach out, I am always happy to chat!

Publications

Quantum initial conditions for curved inflating universes
MI Letey, Z Shumaylov, FJ Agocs, WJ Handley, MP Hobson, AN Lasenby
– Physical Review D
(2024)
109,
123502
Weakly Convex Regularisers for Inverse Problems: Convergence of Critical Points and Primal-Dual Optimisation
Z Shumaylov, J Budd, S Mukherjee, C-B Schönlieb
(2024)
Data-Driven Convex Regularizers for Inverse Problems
S Mukherjee, S Dittmer, Z Shumaylov, S Lunz, O Öktem, CB Schönlieb
– ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
(2024)
13386
Provably Convergent Data-Driven Convex-Nonconvex Regularization
Z Shumaylov, J Budd, S Mukherjee, C-B Schönlieb
(2023)
The Curse of Recursion: Training on Generated Data Makes Models Forget
I Shumailov, Z Shumaylov, Y Zhao, Y Gal, N Papernot, R Anderson
(2023)
Primordial power spectra from k-inflation with curvature
Z Shumaylov, W Handley
– Physical Review D
(2022)
105,
123532
Manipulating SGD with Data Ordering Attacks
I Shumailov, Z Shumaylov, D Kazhdan, Y Zhao, N Papernot, MA Erdogdu, R Anderson
– Advances in Neural Information Processing Systems
(2021)
22,
18021
Learned convex regularizers for inverse problems
S Mukherjee, S Dittmer, Z Shumaylov, S Lunz, O Öktem, C-B Schönlieb
(2020)

Research Group

Mathematics of Information (Applied)

Room

F0.01