Rate My Professor Andrew Eberhard

Eberhard's research focuses on optimization algorithms and theory, numerical and computational mathematics, applied mathematics, pure mathematics, and machine learning. He specializes in the study of convergence and approximation of convex and nonlinear functions and their effects on differential and duality properties of optimization problems. His expertise covers continuous nonsmooth optimization and integer programming, with applications to control theory, nonsmooth optimality conditions, viscosity solutions of elliptic partial differential equations, stochastic optimization with integer variables, duality theory for problem reformulation, the Feasibility Pump for Mixed Integer Programming, progressive hedging convergence analysis, and monotone operator theory alongside convex analysis for control design and signal processing. Key publications include 'Fixed-Time Gradient Dynamics With Time-Varying Coefficients for Continuous-Time Optimization' (2023), 'A study of progressive hedging for stochastic integer programming' (2023), 'Solving Two-stage Quadratic Multiobjective Problems via Optimality and Relaxations' (2024), 'Impact of Mathematical Norms on Convergence of Gradient Descent Algorithms for Deep Neural Networks Learning' (2022), and 'Finding all non-dominated points for a bi-objective generalized assignment problem' (2019). Eberhard has served as Co-chair for optimisation of the Australian Mathematical Society's Mathematics of Computation and Optimisation special interest group from 2018 to 2024 and on the board of the Australian Institute of Mathematical Sciences. He has presented invited talks at events such as the International Conference on Applied Mathematics in Ho Chi Minh City (2024), Variational and Nonsmooth Analysis at The Hong Kong Polytechnic University (2024), Australian-Chinese workshops on optimisation (2004, 2009, 2010, 2013), and as Plenary Speaker at ANZIAM in Warrnambool (2012).