Please use this identifier to cite or link to this item:
Title: Correcting instrumental variation and time-varying drift using parallel and serial multitask learning
Authors: Yan, K
Zhang, D 
Xu, Y
Keywords: Drift correction
Machine olfaction
Multitask learning (MTL)
Transfer learning
Transfer sample
Issue Date: 2017
Publisher: Institute of Electrical and Electronics Engineers
Source: IEEE transactions on instrumentation and measurement, 2017, v. 66, no. 9, p. 2306-2316 How to cite?
Journal: IEEE transactions on instrumentation and measurement 
Abstract: When instruments and sensor systems are used to measure signals, the posterior distribution of test samples often drifts from that of the training ones, which invalidates the initially trained classification or regression models. This may be caused by instrumental variation, sensor aging, and environmental change. We introduce transfer-sample-based multitask learning (TMTL) to address this problem, with a special focus on applications in machine olfaction. Data collected with each device or in each time period define a domain. Transfer samples are the same group of samples measured in every domain. They are used by our method to share knowledge across domains. Two paradigms, parallel and serial transfer, are designed to deal with different types of drift. A dynamic model strategy is proposed to predict samples with known acquisition time. Experiments on three realworld data sets confirm the efficacy of the proposed methods. They achieve good accuracy compared with traditional featurelevel drift correction algorithms and typical labeled-sample-based MTL methods, with few transfer samples needed. TMTL is a practical algorithm framework which can greatly enhance the robustness of sensor systems with complex drift.
ISSN: 0018-9456
EISSN: 1557-9662
DOI: 10.1109/TIM.2017.2707898
Appears in Collections:Journal/Magazine Article

View full-text via PolyU eLinks SFX Query
Show full item record


Last Week
Last month
Citations as of Dec 13, 2018


Last Week
Last month
Citations as of Dec 15, 2018

Page view(s)

Citations as of Dec 9, 2018

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.