MTL-Split: Multi-Task Learning for Edge Devices using Split Computing

Luigi Capogrosso, Enrico Fraccaroli, Samarjit Chakraborty, Franco Fummi, Marco Cristani
7/8/2024
cs.LGcs.CVcs.DC

Official code: intelligolabs/MTL-Split (11 stars, last commit 2024-12-03).

Abstract

Split Computing (SC), where a Deep Neural Network (DNN) is intelligently split with a part of it deployed on an edge device and the rest on a remote server is emerging as a promising approach. It allows the power of DNNs to be leveraged for latency-sensitive applications that do not allow the entire DNN to be deployed remotely, while not having sufficient computation bandwidth available locally. In many such embedded systems scenarios, such as those in the automotive domain, computational resource constraints also necessitate Multi-Task Learning (MTL), where the same DNN is used for multiple inference tasks instead of having dedicated DNNs for each task, which would need more computing bandwidth. However, how to partition such a multi-tasking DNN to be deployed within a SC framework has not been sufficiently studied. This paper studies this problem, and MTL-Split, our novel proposed architecture, shows encouraging results on both synthetic and real-world data. The source code is available at https://github.com/intelligolabs/MTL-Split.

View on arXivView PDF

Code Implementations(1)

intelligolabs/MTL-SplitOfficial100%

Official implementation of the paper "MTL-Split: Multi-Task Learning for Edge Devices using Split Computing" accepted @ DAC 2024.

111Shell, PythonMar 11, 20241 years ago

Similar Papers

A Multi-task Supervised Compression Model for Split Computing

87% match2 repos382025

Enhancing Split Computing and Early Exit Applications through Predefined Sparsity

84% match1 repo52024

Efficient Split Learning LSTM Models for FPGA-based Edge IoT Devices

82% match2025

Split Federated Learning Over Heterogeneous Edge Devices: Algorithm and Optimization

82% match2024

Slimmable Encoders for Flexible Split DNNs in Bandwidth and Resource Constrained IoT Systems

82% match2023

Efficient Parallel Split Learning over Resource-constrained Wireless Edge Networks

82% match2023

Cite this paper

@article{capogrosso2024mtlsplit,
  title  = {MTL-Split: Multi-Task Learning for Edge Devices using Split Computing},
  author = {Luigi Capogrosso and Enrico Fraccaroli and Samarjit Chakraborty and Franco Fummi and Marco Cristani},
  year   = {2024},
  eprint = {2407.05982},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2407.05982v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2407.05982.svg)](https://codeofpaper.com/paper/2407.05982)

Discussion