DARE Featured at Statistical Society of Australia NSW Event

Our team were delighted to be a part of the Statistical Society of Australia‘s (SSA) NSW Branch event in June: Quantifying the known unknowns!

DARE Deputy Director, Professor Lucy Marshall, and PhD candidate, Yiyi Ma, discussed the interaction between statistics and machine learning, aiming to address the lack of interpretability in machine learning.

Lucy shared her research on statistical models for hydrology and explained how environmental research and data analysis can be merged, especially concerning complex data. She highlighted that modern technologies have made the datasets more intricate, introducing uncertainty due to the quality of detection.

The lack of interpretability in machine learning techniques can be addressed by developing hybrid approaches that incorporate machine learning methods within statistical models. As Lucy demonstrated, DARE serves as an excellent Australian example of successfully integrating these hybrid approaches.

Lucy’s studies illustrate the power of data science for future and current modelling applications, but recognise the need for improved training of future engineers and scientists to think probabilistically, and be expert users of data and models.

DARE HDR, Yiyi, introduced her work on particle filters for non-Gaussian models, specifically state-space models, including the significance of discretisation in such algorithms.

This was SSA NSW’s first event at Macquarie University since before COVID and it was great to see many familiar faces in the audience.

Thank you to the SSA NSW Branch for inviting DARE to be a part of your event!