The Methodological Pitfall of Dataset-Driven Research on Deep Learning: An IoT Example

Tianshi Wang, Denizhan Kara, Jinyang Li, Shengzhong Liu, Tarek Abdelzaher, Brian Jalaian

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we highlight a dangerous pitfall in the state-of-the-art evaluation methodology of deep learning algorithms. It results in deceptively good evaluation outcomes on test datasets, whereas the underlying algorithms remain prone to catastrophic failure in practice. We illustrate the pitfall in the context of an Internet-of-Things (IoT) application example and show that it occurs despite the use of cross-validation that breaks down the data into separate training, validation, and testing sets. The pitfall is illustrated by designing two target detection and classification algorithms. One is based on a recently proposed neural network architecture for embedded AI, and the other is based on a traditional machine learning approach with domain-inspired feature engineering. The neural network approach outperforms the traditional one on test data. Yet, it fails in deployment. The mechanics behind the failure are explained and linked to the way the algorithms are trained. Suggestions are presented to avoid the pitfall. The paper is a 'call to arms' to improve the evaluation methodology of machine learning algorithms for mission-critical systems.

Original languageEnglish (US)
Title of host publicationMILCOM 2022 - 2022 IEEE Military Communications Conference
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1082-1087
Number of pages6
ISBN (Electronic)9781665485340
DOIs
StatePublished - 2022
Externally publishedYes
Event2022 IEEE Military Communications Conference, MILCOM 2022 - Rockville, United States
Duration: Nov 28 2022Dec 2 2022

Publication series

NameProceedings - IEEE Military Communications Conference MILCOM
Volume2022-November

Conference

Conference2022 IEEE Military Communications Conference, MILCOM 2022
Country/TerritoryUnited States
CityRockville
Period11/28/2212/2/22

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'The Methodological Pitfall of Dataset-Driven Research on Deep Learning: An IoT Example'. Together they form a unique fingerprint.

Cite this