Recent & Upcoming Talks

2024

Randomised Subspace Methods for Scalable Derivative-Free Optimisation [slides available]

Most algorithms for optimising nonlinear functions rely on access to (possibly stochastic) derivative information. However, for …
19 Sep 2024 UNSW Applied Maths Seminar
Slides

Expected decrease for derivative-free algorithms using random subspaces [slides available]

A promising approach for improving the scalability for DFO methods is to work in low-dimensional subspaces that are iteratively drawn …
28 Jun 2024 DFOS 2024
Slides

Differentiation: the good, the bad, and the ugly [slides available]

For most (sufficiently simple) smooth functions, most high school calculus students can compute its derivative by repeated application …
1 May 2024 Sydney University Mathematical Society
Slides

A Dynamic Linesearch Method for Bilevel Learning [slides available]

16 Feb 2024 SigmaOpt 2024
Slides

2023

Expected decrease for derivative-free algorithms using random subspaces [slides available]

When optimising functions that are black-box, noisy and/or computationally expensive, it may be impractical to get gradient …
11 Dec 2023 WOMBAT/WICO 2023
Slides

Large-scale derivative-free optimization using random subspace methods [slides available]

Many standard optimization algorithms require being able to cheaply and accurately compute derivatives for the objective and/or …
1 Sep 2023 Simons Collaboration on Hidden Symmetries and Fusion Energy
Slides

Analyzing Inexact Hypgergradients for Bilevel Learning [slides available]

Estimating hyperparameters has been a long-standing problem in machine learning. We consider the case where the task at hand is …
1 Jun 2023 SIAM OP23
Slides

2022

Analyzing Inexact Hypergradients for Bilevel Learning [slides available]

Estimating hyperparameters is an important and long-standing problem in machine learning. We consider the case where hyperparameter …
9 Dec 2022 AustMS 2022
Slides

Black-Box Optimisation Techniques for Complex Systems [slides available]

I will introduce some techniques for optimisation suitable for complex systems, such as those involving significant computation or …
29 Nov 2022 Biarri Applied Mathematics Conference
Slides

Large-scale derivative-free optimization using random subspace methods

Many standard optimization algorithms require being able to cheaply and accurately compute derivatives for the objective and/or …
19 Sep 2022 Curtin Centre for Optimisation and Decision Science Colloquium

Large-scale derivative-free optimization using random subspace methods [slides and video available]

Many standard optimization algorithms require being able to cheaply and accurately compute derivatives for the objective and/or …
27 Jul 2022 OPTIMA Seminar Series
Slides Video

2021

Derivative-Free Optimization with Convex Constraints [slides available]

When optimizing functions which are computationally expensive and/or noisy, gradient information is often impractical to obtain or …
14 Dec 2021 WoMBaT 2021
Slides

Derivative-Free Optimization with Convex Constraints [slides available]

When optimizing functions which are computationally expensive and/or noisy, gradient information is often impractical to obtain or …
8 Dec 2021 AustMS 2021
Slides

Inexact Derivative-Free Optimization for Bilevel Learning [slides available]

When variational regularisation methods are used to solve inverse problems, they suffer from the drawback of having potentially many …
26 Oct 2021 INFORMS Annual Meeting 2021
Slides

Inexact Derivative-Free Optimization for Bilevel Learning [slides available]

When variational regularisation methods are used to solve inverse problems, they suffer from the drawback of having potentially many …
14 Oct 2021 Machine Intelligence and Learning Systems Seminar, Université Paris Dauphine-PSL
Slides

Scalable Subspace Methods for Derivative-Free Nonlinear Least-Squares Optimization [slides available]

When optimizing functions which are computationally expensive and/or noisy, gradient information is often impractical to obtain or …
30 Sep 2021 University of Leicester Applied Mathematics Seminar
Slides

Large-Scale Derivative-Free Optimization using Subspace Methods [slides available]

In existing techniques for model-based derivative-free optimization, the computational cost of constructing local models and Lagrange …
20 Jul 2021 SIAM OP21
Slides

Large-Scale Derivative-Free Optimization using Subspace Methods [slides available]

In existing techniques for model-based derivative-free optimization, the computational cost of constructing local models and Lagrange …
12 Jul 2021 EURO 2021
Slides

Large-Scale Derivative-Free Optimization using Subspace Methods [slides available]

In existing techniques for model-based derivative-free optimization, the computational cost of constructing local models and Lagrange …
9 Jul 2021 EUROPT 2021
Slides

Scalable Subspace Methods for Derivative-Free Nonlinear Least-Squares Optimization [slides & video available]

21 Jun 2021 20th IMA Leslie Fox Prize Event
Slides Video

Inexact Derivative-Free Optimization for Bilevel Learning [slides available]

When variational regularisation methods are used to solve inverse problems, they suffer from the drawback of having potentially many …
4 Mar 2021 SIAM CSE 2021
Slides

2020

Block Methods for Scalable Derivative-Free Optimisation [slides available]

Derivative-free optimisation (DFO) methods are an important class of optimisation routines with applications in areas such as in image …
9 Dec 2020 AustMS 2020
Slides

Scalable Derivative-Free Optimization for Nonlinear Least-Squares Problems [slides & video available]

Derivative-free optimisation (DFO) methods are an important class of optimisation routines with applications in areas such as in image …
1 Dec 2020 WoMBaT 2020
Slides Video

Inexact Derivative-Free Optimisation for Bilevel Learning [slides available]

When variational regularisation methods are used to solve inverse problems, they suffer from the drawback of having potentially many …
31 Aug 2020 CTAC 2020
Slides

Derivative-free optimisation for least-squares problems [slides available]

Least-squares problems (such as parameter estimation) are ubiquitous across quantitative disciplines. Optimisation algorithms for …
16 Apr 2020 UNSW Applied Mathematics Seminar
Slides Video

2019

Improving the scalability of model-based derivative-free optimization [slides available]

Derivative-free optimization (DFO) methods are an important class of optimization routines for many problems in data science, such as …
11 Dec 2019 Data Science Down Under
Slides

Improving the efficiency and robustness of black-box optimisation

In classical nonlinear optimisation, the availability of first-order information is crucial to constructing accurate local models for …
28 Nov 2019 ANU Uncertainty Quantification Workshop

Derivative-Free Algorithms for Nonconvex Optimisation

Classical nonconvex optimisation algorithms require the availability of gradient evaluations for constructing local approximations to …
14 Oct 2019 ANU Computational Mathematics Seminar

Improving the scalability of derivative-free optimization for nonlinear least-squares problems

In existing techniques for model-based derivative-free optimization, the computational cost of constructing local models and Lagrange …
6 Aug 2019 ICCOPT 2019

Improving the scalability of derivative-free optimization for nonlinear least-squares problems

In existing techniques for model-based derivative-free optimization, the computational cost of constructing local models and Lagrange …
28 Jun 2019 EUROPT 2019

Improving the scalability of derivative-free optimization for nonlinear least-squares problems

In existing techniques for model-based derivative-free optimization, the computational cost of constructing local models and Lagrange …
26 Jun 2019 28th Biennial Numerical Analysis Conference

Improving the efficiency and robustness of black-box optimisation

In classical nonlinear optimisation, the availability of first-order information is crucial to constructing accurate local models for …
26 Apr 2019 2nd IMA and OR Society Conference on Mathematics of Operational Research

Improving the Flexibility and Robustness of Derivative-Free Optimisation Solvers

Classical nonlinear optimisation algorithms require the availability of gradient evaluations for constructing local approximations to …
8 Mar 2019 Cambridge Image Analysis/Medical Imaging in Healthcare Seminar

2018

Improving the Flexibility and Robustness of Derivative-Free Optimization Solvers

Classical nonlinear optimization algorithms require the availability of gradient evaluations for constructing local approximations to …
30 Aug 2018 National Institute of Informatics Invited Seminar

A flexible, robust and efficient derivative-free solver for least-squares

We present DFO-LS, a software package for derivative-free optimization (DFO) for nonlinear least-squares problems, that has simplified …
4 Jul 2018 ISMP 2018

Improving the efficiency of derivative-free methods for nonlinear least squares problems

We present DFO-GN, a derivative-free version of the Gauss-Newton method for solving nonlinear least-squares problems. As is common in …
27 Jun 2018 6th IMA Conference on Numerical Linear Algebra and Optimization

2017

Derivative-Free Optimisation Methods for Nonlinear Least-Squares Problems

Derivative-free optimisation (DFO) algorithms are a category of optimisation methods for situations when one is unable to compute or …

28 Jun 2017 27th Biennial Numerical Analysis Conference