Federated Learning of Socially Appropriate Agent Behaviours in Simulated Home Environments
This addresses the challenge of ensuring social robots behave appropriately in varied real-world settings, though it is incremental as it adapts existing FL and CL methods to this domain.
The paper tackles the problem of teaching social robots appropriate behaviors in diverse home environments by developing a Federated Learning benchmark that allows robots to learn individually while sharing experiences, and extends it to Federated Continual Learning for incremental learning across contexts, with results showing Federated Averaging as robust and rehearsal-based methods enabling incremental learning.
As social robots become increasingly integrated into daily life, ensuring their behaviours align with social norms is crucial. For their widespread open-world application, it is important to explore Federated Learning (FL) settings where individual robots can learn about their unique environments while also learning from each others' experiences. In this paper, we present a novel FL benchmark that evaluates different strategies, using multi-label regression objectives, where each client individually learns to predict the social appropriateness of different robot actions while also sharing their learning with others. Furthermore, splitting the training data by different contexts such that each client incrementally learns across contexts, we present a novel Federated Continual Learning (FCL) benchmark that adapts FL-based methods to use state-of-the-art Continual Learning (CL) methods to continually learn socially appropriate agent behaviours under different contextual settings. Federated Averaging (FedAvg) of weights emerges as a robust FL strategy while rehearsal-based FCL enables incrementally learning the social appropriateness of robot actions, across contextual splits.