This unit will introduce a wide range of modelling and simulation techniques for tackling real-world problems using a computer. Data is often expensive to obtain, so by harnessing the enormous computational processing power now available to us we can answer what if questions based on data we already have. You will learn how to break a problem down into its key components, identifying necessary assumptions for the purposes of simulation. You will learn how to develop suitable metrics within computational models, to allow comparison of simulation data with real-world data. You will learn how to iteratively improve simulations as you validate them against real results, and you will gain experience in identifying the types of exploratory questions that computational modelling opens up. Programming will be in python. You will learn how to generate probabilistic data, solve systems of differential equations numerically, and tackle complex adaptive systems using agent-based models. Dynamical systems ranging from traffic flow to social segregation will be considered. By doing this unit you will develop the skills to go behind your data, understand why the data you observe might be as it is, and test scenarios which might otherwise be inaccessible. This is an advanced unit. It runs jointly with the associated mainstream unit, however the lab work and assessment requires a greater level of academic rigour. You will be required to engage in more challenging real-world computational modelling problems than the mainstream unit, and explore more deeply the reasons behind simulation results.
Lectures 2x1 hr/wk; Labs 1x1 hr/wk + 1x2 hr/wk
In-lab checkpoints [10%] Assignment [10%] Class test 1 [20%] Class test 2 [20%] Final exam [40%]
HSC Mathematics; DATA1002, or equivalent programming experience, ideally in Python.
48 credit points of 1000 level units with an average of 65Prohibitions
COSC1003 or COSC1903 or COSC2002