Input quality aware convolutional LSTM networks for virtual marine sensors
Research output: Contribution to journal › Journal article › Research › peer-review
Standard
Input quality aware convolutional LSTM networks for virtual marine sensors. / Oehmcke, Stefan; Zielinski, Oliver; Kramer, Oliver.
In: Neurocomputing, Vol. 275, 31.01.2018, p. 2603-2615.Research output: Contribution to journal › Journal article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Input quality aware convolutional LSTM networks for virtual marine sensors
AU - Oehmcke, Stefan
AU - Zielinski, Oliver
AU - Kramer, Oliver
PY - 2018/1/31
Y1 - 2018/1/31
N2 - The harsh environmental conditions in a marine area make continuous observations of it challenging. To temporally or permanently replace faulty hardware sensors, reliable virtual sensors are getting more and more important. This paper introduces a deep learning architecture for a marine virtual sensor application that is able to utilize input quality information. We propose a novel input quality based dropout layer to take advantage of Information about the quality of the input sensors. The virtual sensor models are built upon convolutional and recurrent long short-term memory layers. We apply a time dimensionality reduction method called exPAA that retains finer details from recent values, but past information with less details are also available to the model. As interpretable reliability is important virtual sensors, we include predictive uncertainty based on dropout and Monte Carlo predictions into our neural network. Experimental results show that we perform better than the baseline and our input quality based dropout layer improves on these results even further. We also provide insights on the learned uncertainty intervals as well as the utilized linear and non-linear correlations of the first convolutional layer.
AB - The harsh environmental conditions in a marine area make continuous observations of it challenging. To temporally or permanently replace faulty hardware sensors, reliable virtual sensors are getting more and more important. This paper introduces a deep learning architecture for a marine virtual sensor application that is able to utilize input quality information. We propose a novel input quality based dropout layer to take advantage of Information about the quality of the input sensors. The virtual sensor models are built upon convolutional and recurrent long short-term memory layers. We apply a time dimensionality reduction method called exPAA that retains finer details from recent values, but past information with less details are also available to the model. As interpretable reliability is important virtual sensors, we include predictive uncertainty based on dropout and Monte Carlo predictions into our neural network. Experimental results show that we perform better than the baseline and our input quality based dropout layer improves on these results even further. We also provide insights on the learned uncertainty intervals as well as the utilized linear and non-linear correlations of the first convolutional layer.
KW - Coastal observatory
KW - Convolutional LSTM
KW - Dropout
KW - Time series
KW - Uncertainty prediction
KW - Virtual sensor
UR - http://www.scopus.com/inward/record.url?scp=85036617467&partnerID=8YFLogxK
U2 - 10.1016/j.neucom.2017.11.027
DO - 10.1016/j.neucom.2017.11.027
M3 - Journal article
AN - SCOPUS:85036617467
VL - 275
SP - 2603
EP - 2615
JO - Neurocomputing
JF - Neurocomputing
SN - 0925-2312
ER -
ID: 223196075