Federated Learning with Uncertainty via Distilled Predictive Distributions

no code implementations • 15 Jun 2022 • Shrey Bhatt, Aishwarya Gupta, Piyush Rai

In many situations, however, especially in limited data settings, it is beneficial to take into account the uncertainty in the model parameters at each client as it leads to more accurate predictions and also because reliable estimates of uncertainty can be used for tasks, such as out-of-distribution (OOD) detection, and sequential decision-making tasks, such as active learning.

Active Learning Decision Making +2