TY - JOUR
T1 - InceptionTime
T2 - Finding AlexNet for time series classification
AU - Ismail Fawaz, Hassan
AU - Lucas, Benjamin
AU - Forestier, Germain
AU - Pelletier, Charlotte
AU - Schmidt, Daniel F.
AU - Weber, Jonathan
AU - Webb, Geoffrey I.
AU - Idoumghar, Lhassane
AU - Muller, Pierre Alain
AU - Petitjean, François
N1 - Publisher Copyright:
© 2020, The Author(s), under exclusive licence to Springer Science+Business Media LLC, part of Springer Nature.
PY - 2020/11/1
Y1 - 2020/11/1
N2 - This paper brings deep learning at the forefront of research into time series classification (TSC). TSC is the area of machine learning tasked with the categorization (or labelling) of time series. The last few decades of work in this area have led to significant progress in the accuracy of classifiers, with the state of the art now represented by the HIVE-COTE algorithm. While extremely accurate, HIVE-COTE cannot be applied to many real-world datasets because of its high training time complexity in O(N2· T4) for a dataset with N time series of length T. For example, it takes HIVE-COTE more than 8 days to learn from a small dataset with N= 1500 time series of short length T= 46. Meanwhile deep learning has received enormous attention because of its high accuracy and scalability. Recent approaches to deep learning for TSC have been scalable, but less accurate than HIVE-COTE. We introduce InceptionTime—an ensemble of deep Convolutional Neural Network models, inspired by the Inception-v4 architecture. Our experiments show that InceptionTime is on par with HIVE-COTE in terms of accuracy while being much more scalable: not only can it learn from 1500 time series in one hour but it can also learn from 8M time series in 13 h, a quantity of data that is fully out of reach of HIVE-COTE.
AB - This paper brings deep learning at the forefront of research into time series classification (TSC). TSC is the area of machine learning tasked with the categorization (or labelling) of time series. The last few decades of work in this area have led to significant progress in the accuracy of classifiers, with the state of the art now represented by the HIVE-COTE algorithm. While extremely accurate, HIVE-COTE cannot be applied to many real-world datasets because of its high training time complexity in O(N2· T4) for a dataset with N time series of length T. For example, it takes HIVE-COTE more than 8 days to learn from a small dataset with N= 1500 time series of short length T= 46. Meanwhile deep learning has received enormous attention because of its high accuracy and scalability. Recent approaches to deep learning for TSC have been scalable, but less accurate than HIVE-COTE. We introduce InceptionTime—an ensemble of deep Convolutional Neural Network models, inspired by the Inception-v4 architecture. Our experiments show that InceptionTime is on par with HIVE-COTE in terms of accuracy while being much more scalable: not only can it learn from 1500 time series in one hour but it can also learn from 8M time series in 13 h, a quantity of data that is fully out of reach of HIVE-COTE.
KW - Deep learning
KW - Inception
KW - Scalable model
KW - Time series classification
UR - http://www.scopus.com/inward/record.url?scp=85090433635&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85090433635&partnerID=8YFLogxK
U2 - 10.1007/s10618-020-00710-y
DO - 10.1007/s10618-020-00710-y
M3 - Article
AN - SCOPUS:85090433635
SN - 1384-5810
VL - 34
SP - 1936
EP - 1962
JO - Data Mining and Knowledge Discovery
JF - Data Mining and Knowledge Discovery
IS - 6
ER -