OCTOPUS: Overcoming Performance andPrivatization Bottlenecks in Distributed Learning

TL;DR

This paper proposes a novel distributed learning scheme that reduces communication costs and enhances data privacy by encoding local data into latent representations, enabling efficient multi-task learning from non-iid data sources.

Contribution

It introduces a new collaborative learning approach combining latent compression and privatization strategies, addressing communication and privacy challenges in federated learning.

Findings

Achieves comparable accuracy to centralized learning on image and speech datasets.

Reduces communication overhead through feature encoding and latent representations.

Ensures privacy of local data via disentanglement and privatization strategies.

Abstract

The diversity and quantity of data warehouses, gathering data from distributed devices such as mobile devices, can enhance the success and robustness of machine learning algorithms. Federated learning enables distributed participants to collaboratively learn a commonly-shared model while holding data locally. However, it is also faced with expensive communication and limitations due to the heterogeneity of distributed data sources and lack of access to global data. In this paper, we investigate a practical distributed learning scenario where multiple downstream tasks (e.g., classifiers) could be efficiently learned from dynamically-updated and non-iid distributed data sources while providing local data privatization. We introduce a new distributed/collaborative learning scheme to address communication overhead via latent compression, leveraging global data while providing privatization of local data without additional cost due to encryption or perturbation. This scheme divides learning into (1) informative feature encoding, and transmitting the latent representation of local data to address communication overhead; (2) downstream tasks centralized at the server using the encoded codes gathered from each node to address computing overhead. Besides, a disentanglement strategy is applied to address the privatization of sensitive components of local data. Extensive experiments are conducted on image and speech datasets. The results demonstrate that downstream tasks on the compact latent representations with the privatization of local data can achieve comparable accuracy to centralized learning.