PhD Seminar: Stochastic approximation for target tracking and mine planning optimisation
by Kim Levy
Abstract: In the first part, I introduce the problem of real-time tracking of a moving
target, such as a signal, through noisy measurements. I show how combining
change-point monitoring with stochastic approximation can improve the estimates
of the target in a particular non-stationary environment where abrupt changes
occur suddenly and unpredictably. I will also present my latest idea: shadow tracking.
In the second part, I address the problem of optimizing mining infrastructure to
improve strategic managerial decisions. Common optimization methods are
impractical for realistic size models. We address the curse of dimensionality via
threshold optimization using stochastic approximation.
For More Information: Contact Paul Pearce (P.Pearce@ms.unimelb.edu.au) or Paul Norbury (P.Norbury@ms.unimelb.edu.au)