FedVLN: Privacy-preserving Federated Vision-and-Language Navigation

TL;DR

This paper introduces FedVLN, a federated learning framework for Vision-and-Language Navigation that enhances privacy by keeping environment data local while maintaining competitive navigation performance.

Contribution

The paper proposes a novel federated learning approach for VLN that ensures data privacy during training and exploration, with a decentralized training strategy and federated pre-exploration.

Findings

Decentralized VLN models achieve comparable results to centralized training.

Federated pre-exploration outperforms centralized pre-exploration.

The framework effectively preserves privacy of unseen environments.

Abstract

Data privacy is a central problem for embodied agents that can perceive the environment, communicate with humans, and act in the real world. While helping humans complete tasks, the agent may observe and process sensitive information of users, such as house environments, human activities, etc. In this work, we introduce privacy-preserving embodied agent learning for the task of Vision-and-Language Navigation (VLN), where an embodied agent navigates house environments by following natural language instructions. We view each house environment as a local client, which shares nothing other than local updates with the cloud server and other clients, and propose a novel federated vision-and-language navigation (FedVLN) framework to protect data privacy during both training and pre-exploration. Particularly, we propose a decentralized training strategy to limit the data of each client to its local model training and a federated pre-exploration method to do partial model aggregation to improve model generalizability to unseen environments. Extensive results on R2R and RxR datasets show that under our FedVLN framework, decentralized VLN models achieve comparable results with centralized training while protecting seen environment privacy, and federated pre-exploration significantly outperforms centralized pre-exploration while preserving unseen environment privacy.