Federated Continual Learning for Socially Aware Robotics

TL;DR

This paper introduces a decentralized federated continual learning approach for social robots, enhancing privacy and personalization by enabling robots to learn from distributed interactions without centralized data collection.

Contribution

It proposes a novel Elastic Transfer algorithm that combines federated and continual learning to improve privacy and adaptability in social robots.

Findings

Decentralized learning is feasible for social navigation tasks.

Elastic Transfer improves parameter relevance preservation.

The approach enhances privacy and personalization in social robots.

Abstract

From learning assistance to companionship, social robots promise to enhance many aspects of daily life. However, social robots have not seen widespread adoption, in part because (1) they do not adapt their behavior to new users, and (2) they do not provide sufficient privacy protections. Centralized learning, whereby robots develop skills by gathering data on a server, contributes to these limitations by preventing online learning of new experiences and requiring storage of privacy-sensitive data. In this work, we propose a decentralized learning alternative that improves the privacy and personalization of social robots. We combine two machine learning approaches, Federated Learning and Continual Learning, to capture interaction dynamics distributed physically across robots and temporally across repeated robot encounters. We define a set of criteria that should be balanced in decentralized robot learning scenarios. We also develop a new algorithm -- Elastic Transfer -- that leverages importance-based regularization to preserve relevant parameters across robots and interactions with multiple humans. We show that decentralized learning is a viable alternative to centralized learning in a proof-of-concept Socially-Aware Navigation domain, and demonstrate how Elastic Transfer improves several of the proposed criteria.