Adaptive Collaborative Learning with Data Silos

The School of EECS is hosting the following Seminar:

Adaptive Collaborative Learning with Data Silos

Speaker: Achmad Ginanjar
Host: Prof Xue Li

Abstract: The data silos environment is a common challenge, especially in sectors like government that involve numerous business processes. Within this context, several difficulties arise when building a collaborative machine learning model. These challenges include issues like collaborative learning, safeguarding privacy, ensuring data availability, accommodating multiple models, and addressing concerns related to political correctness. Take, for example, the Indonesian Tax Office, which is currently developing a compliance risk management (CRM) model. Within the CRM initiative, they encounter the above data silos-related issues. This study investigates strategies to effectively handle these challenges related to data silos-related challenges effectively. The goal is to develop a solution that can effectively address data scarcity, preservation of privacy, and integration of multiple models. Furthermore, since some of these challenges are politically sensitive, the final model is expected to offer flexibility in addressing these concerns..

Bio: Achmad Ginanjar received a Master of Data Science degree from Monash University, Australia in 2019. He is currently working toward a Ph.D. degree with the Electrical Engineering Department at the University of Queensland (UQ), Brisbane, QLD, Australia. Before joining UQ, he is an IT professional from The Indonesia Tax Office and a data analytic lecture at Indonesia State College of Accountancy. His current research interests include machine learning, optimalisation, parallel algorithm.