Deakin University
Browse

File(s) under permanent embargo

Extending deep learning to new classes without retraining

Version 2 2024-06-03, 12:33
Version 1 2020-08-06, 09:46
conference contribution
posted on 2024-06-03, 12:33 authored by J Schulz, C Veal, A Buck, D Anderson, J Keller, M Popescu, G Scott, DKC Ho, Tim WilkinTim Wilkin
The focus of this article is extending classifiers from N classes to N+1 classes without retraining for tasks like explosive hazard detection (EHD) and automatic target recognition (ATR). In recent years, deep learning has become state-of-the-art across domains. However, algorithms like convolutional neural networks (CNNs) suffer from the assumption of a closed-world model. That is, once a model is learned, a new class cannot usually be added without changes in the architecture and retraining. Herein, we put forth a way to extend a number of deep learning algorithms while keeping their features in a locked state; i.e., features are not retrained for the new N+1 class. Different feature transformations, metrics, and classifiers are explored to assess the degree to which a new sample belongs to one of the N classes and a decision rule is used for classification. Whereas this extends a deep learner, it does not tell us if a network with locked features has the potential to be extended. Therefore, we put forth a new method based on visually assessing cluster tendency to assess the degree to which a deep learner can be extended (or not). Lastly, while we are primarily focused on tasks like aerial EHD and ATR, experiments herein are for benchmark community data sets for sake of reproducible research.

History

Volume

11418

Pagination

1141803-1-1141803-14

Location

Online Only, United States

Start date

2020-04-27

End date

2020-05-08

ISSN

0277-786X

eISSN

1996-756X

ISBN-13

9781510636132

Language

eng

Publication classification

E1 Full written paper - refereed

Editor/Contributor(s)

Bishop SS, Isaacs JC

Title of proceedings

Proceedings of SPIE Detection and Sensing of Mines, Explosive Objects, and Obscured Targets XXV

Event

Society of Photo-Optical Instrumentation Engineers. Conference (2020 : Online Only, United States)

Publisher

SPIE

Place of publication

Washington, D.C.

Series

Society of Photo-Optical Instrumentation Engineers Conference