Improving Robustness of Spectrogram Classifiers with Neural Stochastic Differential Equations

Joel Brogan, Olivera Kotevska, Anibely Torres, Sumit Jha, Mark Adams

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Signal analysis and classification is fraught with high levels of noise and perturbation. Computer-vision-based deep learning models applied to spectrograms have proven useful in the field of signal classification and detection; however, these methods aren't designed to handle the low signal-to-noise ratios inherent within non-vision signal processing tasks. While they are powerful, they are currently not the method of choice in the inherently noisy and dynamic critical infrastructure domain, such as smart-grid sensing, anomaly detection, and non-intrusive load monitoring. Currently, these models can be brittle, which makes them susceptible to noisy input. This also means they have sub-optimal stability of explanation outputs. Experts and technicians using these models to make decisions in real world scenarios need assurance that a model is performing as it is supposed to. The classification or prediction outputs it generates should be sound and grounded, not likely to change in the presence of shifting noise landscapes. In this work, we explore the idea of Neural Stochastic Differential Equations (NSDE's) to improve the robustness of models trained to classify time series data and the effect of NSDE's on the explainability of outputs. We then test the effectiveness of these approaches by applying them to a non-intrusive load monitoring (NILM) dataset that consists of simulated harmonic signals injected into a real building.

Original languageEnglish
Title of host publication34th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2024 - Proceedings
PublisherIEEE Computer Society
ISBN (Electronic)9798350372250
DOIs
StatePublished - 2024
Event34th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2024 - London, United Kingdom
Duration: Sep 22 2024Sep 25 2024

Publication series

NameIEEE International Workshop on Machine Learning for Signal Processing, MLSP
ISSN (Print)2161-0363
ISSN (Electronic)2161-0371

Conference

Conference34th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2024
Country/TerritoryUnited Kingdom
CityLondon
Period09/22/2409/25/24

Keywords

  • robustness
  • signal classification
  • XAI

Fingerprint

Dive into the research topics of 'Improving Robustness of Spectrogram Classifiers with Neural Stochastic Differential Equations'. Together they form a unique fingerprint.

Cite this