TabVFL: Improving Latent Representation in Vertical Federated Learning

TL;DR

TabVFL introduces a novel federated learning framework that enhances latent feature representations, preserves feature correlations, and improves robustness in vertical federated learning on tabular data.

Contribution

It proposes a new distributed framework that jointly learns latent representations, maintains feature correlations, and enhances robustness against client failures in VFL.

Findings

Outperforms prior methods with 26.12% F1-score improvement.

Effectively preserves feature correlations across participants.

Provides robustness against client failures during training.

Abstract

Autoencoders are popular neural networks that are able to compress high dimensional data to extract relevant latent information. TabNet is a state-of-the-art neural network model designed for tabular data that utilizes an autoencoder architecture for training. Vertical Federated Learning (VFL) is an emerging distributed machine learning paradigm that allows multiple parties to train a model collaboratively on vertically partitioned data while maintaining data privacy. The existing design of training autoencoders in VFL is to train a separate autoencoder in each participant and aggregate the latent representation later. This design could potentially break important correlations between feature data of participating parties, as each autoencoder is trained on locally available features while disregarding the features of others. In addition, traditional autoencoders are not specifically designed for tabular data, which is ubiquitous in VFL settings. Moreover, the impact of client failures during training on the model robustness is under-researched in the VFL scene. In this paper, we propose TabVFL, a distributed framework designed to improve latent representation learning using the joint features of participants. The framework (i) preserves privacy by mitigating potential data leakage with the addition of a fully-connected layer, (ii) conserves feature correlations by learning one latent representation vector, and (iii) provides enhanced robustness against client failures during training phase. Extensive experiments on five classification datasets show that TabVFL can outperform the prior work design, with 26.12% of improvement on f1-score.