# Distributed Layer-Partitioned Training for Privacy-Preserved Deep Learning

## TL;DR

This paper introduces a distributed layer-partitioned training method with step-wise activation functions to enhance privacy preservation in deep learning, enabling secure training on sensitive data.

## Contribution

It proposes a novel distributed training approach that partitions layers and uses step-wise activations to protect sensitive information during model training.

## Key findings

- Method is simple and effective
- Preserves privacy during distributed training
- Achieves comparable accuracy to traditional methods

## Abstract

Deep Learning techniques have achieved remarkable results in many domains. Often, training deep learning models requires large datasets, which may require sensitive information to be uploaded to the cloud to accelerate training. To adequately protect sensitive information, we propose distributed layer-partitioned training with step-wise activation functions for privacy-preserving deep learning. Experimental results attest our method to be simple and effective.