# BottleNet: A Deep Learning Architecture for Intelligent Mobile Cloud Computing Services

## TL;DR

BottleNet is a novel deep learning architecture that reduces latency and energy consumption in mobile cloud computing by compressing features sent to the cloud, with a training method to maintain accuracy.

## Contribution

The paper introduces BottleNet, a new architecture and training method that significantly improves mobile cloud computing efficiency while preserving model accuracy.

## Key findings

- 30x reduction in end-to-end latency
- 40x decrease in mobile energy consumption
- Negligible accuracy loss

## Abstract

Recent studies have shown the latency and energy consumption of deep neural networks can be significantly improved by splitting the network between the mobile device and cloud. This paper introduces a new deep learning architecture, called BottleNet, for reducing the feature size needed to be sent to the cloud. Furthermore, we propose a training method for compensating for the potential accuracy loss due to the lossy compression of features before transmitting them to the cloud. BottleNet achieves on average 30x improvement in end-to-end latency and 40x improvement in mobile energy consumption compared to the cloud-only approach with negligible accuracy loss.