AI/ML methods are revolutionizing scientific discovery. In this project we invite you to build novel AI/Ml Algorithms and apply them to a range of challenging science and engineering domains. Example application areas in the area of climate and weather forecasting include, but are not limited to the following: -
- Ocean SST forecasts
- Data driven weather forecasting
- Marine heatwaves
- Future state of the Great Barrier Reef
- Extreme weather events
Develop new ML optimization algorithms and explore the fundamentals of applying ML/AI methods to scientific problems including for example:
- apply Levenburg-Marquadt optimisation to ML models at scale on HPC systems
- Double descent is a surprising phenomenon in machine learning, in which as the number of model parameters grows relative to the number of data, test error drops as models grow ever larger into the highly overparameterized (data undersampled) regime.
- Improving our understanding AI extrapolation e.g. knowing when it is occurring is essential
HPC is a key tool that we will use in these projects.