Fred Roosta   Professor School of Mathematics and Physics, University of Queensland, Australia Email: fred.roosta@uq.edu.au Office: Priestley Building (67) - Room 754 Phone: +61 7 336 53259

Preprints and Publications

2025

• Determinant Estimation under Memory Constraints and Neural Scaling Laws
  Siavash Ameli, Chris van der Heide, Liam Hodgkinson, Fred Roosta, Michael W. Mahoney
  Technical Report, Preprint, 2025.

• First-ish Order Methods: Hessian-aware Scalings of Gradient Descent
  Oscar Smee, Fred Roosta, Stephen J. Wright
  Technical Report, Preprint, 2025.

• Uncertainty Quantification with the Empirical Neural Tangent Kernel
  Joseph Wilson, Chris van der Heide, Liam Hodgkinson, Fred Roosta
  Technical Report, Preprint, 2025.

• Complexity Guarantees for Nonconvex Newton-MR Under Inexact Hessian Information
  Alexander Lim, Fred Roosta
  IMA Journal of Numerical Analysis (IMAJNA), 2025.

• Training-free Medical Image Inversion via Bi-level Guided Diffusion Models
  Hossein Askari, Fred Roosta, Hongfu Sun
  IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2025.

2024

• Conjugate Direction Methods Under Inconsistent Systems
  Alexander Lim, Yang Liu, Fred Roosta
  Technical Report, Preprint, 2024.

• SALSA: Sequential Approximate Leverage-Score Algorithm with Application in Analyzing Big Time Series Data

  Ali Eshragh, Luke Yerbury, Asef Nazari, Fred Roosta, Michael W. Mahoney
  Technical Report, Preprint, 2024.

• Inexact Newton-type Methods for Optimisation with Nonnegativity Constraints
  Oscar Smee, Fred Roosta
  Proc. of International Conference on Machine Learning (ICML), 2024. [Code]

• Manifold Integrated Gradients: Riemannian Geometry for Feature Attribution
  Eslam Zaher, Maciej Trzaskowski, Quan Nguyen, Fred Roosta
  Proc. of International Conference on Machine Learning (ICML), 2024. [Code]

• Non-Uniform Smoothness for Gradient Descent
  Albert S. Berahas, Lindon Roberts, Fred Roosta
  Transactions on Machine Learning Research (TMLR), 2024.

2023

• A PAC-Bayesian Perspective on the Interpolating Information Criterion
  Liam Hodgkinson, Chris van der Heide, Robert Salomone, Fred Roosta, Michael W. Mahoney
  NeurIPS 2023 Workshop on Mathematics of Modern Machine Learning.

• Obtaining Pseudo-inverse Solutions With MINRES
  Yang Liu, Andre Milzarek, Fred Roosta
  Technical Report, Preprint, 2023.

• The Interpolating Information Criterion for Overparameterized Models
  Liam Hodgkinson, Chris van der Heide, Robert Salomone, Fred Roosta, Michael W. Mahoney
  Technical Report, Preprint, 2023.

• STimage: robust, confident and interpretable models for predicting gene markers from cancer histopathological images
  Xiao Tan, Onkar Mulay, Samual MacDonald, Taehyun Kim, Jason Werry, Peter T Simpson, Fred Roosta, Maciej Trzaskowski, Quan Nguyen
  Technical Report, Preprint, 2023.

• Inexact Newton-CG Algorithms With Complexity Guarantees
  Zhewei Yao, Peng Xu, Fred Roosta, Stephen J. Wright, and Michael W. Mahoney
  IMA Journal of Numerical Analysis (IMAJNA), 2023.

• Monotonicity and Double Descent in Uncertainty Estimation with Gaussian Processes
  Liam Hodgkinson, Chris van der Heide, Fred Roosta, Michael W. Mahoney
  Proc. of International Conference on Machine Learning (ICML), 2023.

• Generalising uncertainty improves accuracy and safety of deep learning analytics applied to oncology
  Samual MacDonald, Helena Foley, Melvyn Yap, Rebecca L Johnston, Kaiah Steven, Lambros T Koufariotis, Sowmya Sharma, Scott Wood, Venkateswar Addala, John Pearson, Fred Roosta, Nicola Waddell, Olga Kondrashova, Maciej Trzaskowski
  Scientific Reports, 2023.

2022

• A Newton-MR algorithm with complexity guarantees for nonconvex smooth unconstrained optimization
  Yang Liu, Fred Roosta
  Technical Report, Preprint, 2022. [code]

• Newton-MR: Inexact Newton Method With Minimum Residual Sub-problem
  Fred Roosta, Yang Liu, Peng Xu, Michael W. Mahoney
  EURO Journal on Computational Optimization (EJCO), 2022. [code]

• MINRES: From Negative Curvature Detection to Monotonicity Properties
  Yang Liu, Fred Roosta
  SIAM Journal on Optimization (SIOPT), 2022.

• LSAR: Efficient Leverage Score Sampling Algorithm for the Analysis of Big Time Series Data

  Ali Eshragh, Fred Roosta, Asef Nazari, Michael W. Mahoney
  Journal of Machine Learning Research (JMLR), 2022.

• Confirming the Lassonde Curve through life cycle analysis and its effect on share price: A case study of three ASX listed gold companies.

  Timothy Rijsdijk, Micah Nehring, Mehmet Kizil, and Fred Roosta
  Resources Policy, 2022.

2021

• Invexifying Regularization of Non-Linear Least-Squares Problems
  Rixon Crane, Fred Roosta
  Technical Report, Preprint, 2021. [code]

• Evolution and application of digital technologies to predict crop type and crop phenology in agriculture
  A B Potgieter, Yan Zhao, Pablo J Zarco-Tejada, Karine Chenu, Yifan Zhang, Kenton Porker, Ben Biddulph, Yash P Dang, Tim Neale, Fred Roosta, Scott Chapman
  in silico Plants, Oxford University Press, 2021.

• Non-PSD Matrix Sketching with Applications to Regression and Optimization
  Zhili Feng, Fred Roosta, David Woodruff
  Proc. of the Conference on Uncertainty in Artificial Intelligence (UAI), 2021.

• Stochastic Continuous Normalizing Flows: Training SDEs as ODEs
  Liam Hodgkinson, Chris van der Heide, Fred Roosta, Michael W. Mahoney
  Proc. of the Conference on Uncertainty in Artificial Intelligence (UAI), 2021.

• Shadow Manifold Hamiltonian Monte Carlo
  Chris van der Heide, Liam Hodgkinson, Fred Roosta, Dirk Kroese
  Proc. of the International Conference on Artificial Intelligence and Statisticst (AISTATS), 2021.

• Inexact Non-Convex Newton-Type Methods
  Zhewei Yao, Peng Xu, Fred Roosta, Michael W. Mahoney
  INFORMS Journal on Optimization, 2021.

• Convergence of Newton-MR under Inexact Hessian Information
  Yang Liu, Fred Roosta
  SIAM Journal on Optimization (SIOPT), 2021.

2020