People_
Dr Lindon Roberts
Dr
Phone
02 9351 2973
Address
F07 - Carslaw Building
The University of Sydney
Lindon Roberts is a lecturer in the School of Mathematics and Statistics. He completed his DPhil (Doctor of Philosophy) at the University of Oxford and was previously an MSI (Mathematical Science Institute) Fellow at the Australian National University. His research interests are in numerical optimisation, including nonconvex, derivative-free and stochastic optimisation, and applications in machine learning. He currently teaches for MATH2070 (Optimisation and Financial Mathematics) and FMAT3888 (Projects in Financial Mathematics).
Timetable
- Algorithms for convex-constrained derivative-free optimisation
- Scalable derivative-free optimisation using random subspace methods
- Bilevel learning for image analysis
- Adaptive smoothness in nonconvex optimisation
- Optimising illumination patterns for ghost imaging
- Member of SIAM
- Member of AustMS
- Best Paper Award, Mathematical Programming Computation (2019)
- Reddick Prize, Mathematical Institute, University of Oxford (2020)
- IMA Leslie Fox Prize for Numerical Analysis (2021)
| Project title | Research student |
|---|---|
| Tensor Methods for Sequential Scientific Machine Learning with Application to Digital Twins | Daniel TRAN |
Publications
Expand all
Journals
- Hare, W., Roberts, L., Royer, C. (2025). Expected decrease for derivative-free algorithms using random subspaces. Mathematics of Computation, 94(351), 277-304. [More Information]
- Roberts, L. (2025). Model Construction for Convex-Constrained Derivative-Free Optimization. SIAM Journal on Optimization, 35(2), 622-650. [More Information]
- Ehrhardt, M., Roberts, L. (2024). Analyzing inexact hypergradients for bilevel learning. IMA Journal of Applied Mathematics, 89(1), 254-278. [More Information]
show 15 more
Expand all
2025
- Hare, W., Roberts, L., Royer, C. (2025). Expected decrease for derivative-free algorithms using random subspaces. Mathematics of Computation, 94(351), 277-304. [More Information]
- Roberts, L. (2025). Model Construction for Convex-Constrained Derivative-Free Optimization. SIAM Journal on Optimization, 35(2), 622-650. [More Information]
2024
- Ehrhardt, M., Roberts, L. (2024). Analyzing inexact hypergradients for bilevel learning. IMA Journal of Applied Mathematics, 89(1), 254-278. [More Information]
- Berahas, A., Roberts, L., Roosta-Khorasani, F. (2024). Non-uniform smoothness for gradient descent. Transactions on Machine Learning Research, 2, Article 29-20 pages.
2023
- Roberts, L., Royer, C. (2023). Direct search based on probabilistic descent in reduced spaces. SIAM Journal on Optimization, 33(4), 3057-3082. [More Information]
- Aminzadeh, A., Roberts, L., Young, B., Chiang, C., Svalbe, I., Paganin, D., Kingston, A. (2023). Mask design, fabrication, and experimental ghost imaging applications for patterned X-ray illumination. Optics Express, 31(15), 24328-24346. [More Information]
- Kingston, A., Aminzadeh, A., Roberts, L., Pelliccia, D., Svalbe, I., Peganin, D. (2023). Optimizing nonconfigurable, transversely displaced masks for illumination patterns in classical ghost imaging. Physical Review A, 107(February 2023), 023524-1-023524-24. [More Information]
2022
- Roberts, L., Smyth, E. (2022). A simplified convergence theory for Byzantine resilient stochastic gradient descent. Euro Journal on Computational Optimization, 10, 1-18. [More Information]
- Tett, S., Gregory, J., Freychet, N., Cartis, C., Mineter, M., Roberts, L. (2022). Does Model Calibration Reduce Uncertainty in Climate Projections? Journal of Climate, 35(8), 2585-2602. [More Information]
- Cartis, C., Roberts, L., Sheridan-Methven, O. (2022). Escaping local minima with local derivative-free methods: a numerical investigation. Optimization, 71(8), 2343-2373. [More Information]
show 4 more
2021
- Ehrhardt, M., Roberts, L. (2021). Inexact Derivative-Free Optimization for Bilevel Learning. Journal of Mathematical Imaging and Vision, 63(5), 580-600. [More Information]
2019
- Cartis, C., Roberts, L. (2019). A derivative-free Gauss–Newton method. Mathematical Programming Computation, 11(4), 631-674. [More Information]
- Cartis, C., Fiala, J., Marteau, B., Roberts, L. (2019). Improving the flexibility and robustness of model-based derivative-free optimization solvers. ACM Transactions on Mathematical Software, 45(3), 1-35. [More Information]
2017
- Roberts, L., Nordgard-Hansen, E., Mikkelsen, O., Halvorsen, S., Van Gorder, R. (2017). A heat and mass transfer study of carbon paste baking. International Communications in Heat and Mass Transfer, 88, 9-19. [More Information]
Selected Grants
2024
- Robust Derivative-Free Algorithms for Complex Optimisation Problems, Roberts L, Australian Research Council (ARC)/Discovery Early Career Researcher Award (DECRA)
2022
- Theory and Algorithms for Numerical Optimisation, Roberts L, Faculty of Science/Faculty Startup Scheme