Kanish is deeply passionate about exploring the potential of artificial intelligence and machine learning. Therefore his Phd research is focused on improving the utility and privacy of Federated Learning (FL) through the application of differential privacy (DP) techniques. The specific area of improvement is DP-SGD and DP-FedAvg algorithms, and the research aims to propose a novel modification that uses a Haar wavelet transform technique to provide a better utility-to-privacy tradeoff than the vanilla algorithms.
For more information, please contact one of our APR.Intern Business Development Managers:
- Justin Mabbutt (VIC & TAS) – j.mabbutt@aprintern.org.au / 0413 050 952
- Mark Ovens (NSW & ACT) – m.ovens@aprintern.org.au / 0400 764 763
- Glen Sheldon – g.sheldon@aprintern.org.au / 0431 832 788
- Stacey Hansen – stacey.hansen@aprintern.org.au / 0438 098 399
