| Paper ID | MLSP-29.3 |
| Paper Title |
BLEND-RES^2NET: BLENDED REPRESENTATION SPACE BY TRANSFORMATION OF RESIDUAL MAPPING WITH RESTRAINED LEARNING FOR TIME SERIES CLASSIFICATION |
| Authors |
Arijit Ukil, Tata Consultancy Services, India; Antonio J. Jara, HOP Ubiquitous (HOPU), Spain; Leandro Marin, University of Murcia, Spain |
| Session | MLSP-29: Deep Learning for Time Series |
| Location | Gather.Town |
| Session Time: | Thursday, 10 June, 14:00 - 14:45 |
| Presentation Time: | Thursday, 10 June, 14:00 - 14:45 |
| Presentation |
Poster
|
| Topic |
Machine Learning for Signal Processing: [MLR-DEEP] Deep learning techniques |
| IEEE Xplore Open Preview |
Click here to view in IEEE Xplore |
| Virtual Presentation |
Click here to watch in the Virtual Conference |
| Abstract |
The typical problem like insufficient training instances in time series classification task demands novel deep neural network architecture to warrant consistent and accurate performance. Deep Residual Network (ResNet) learns through H(x)=F(x)+x, where F(x) is a nonlinear function. We propose Blend-Res2Net that blends two different representation spaces: H^1 (x)=F(x)+Trans(x) and H^2 (x)=F(Trans(x))+x with the intention of learning over richer representation by capturing the temporal as well as the spectral signatures (Trans(∙) represents the transformation function). The sophistication of richer representation for better understanding of the complex structure of time series signals, may result in higher generalization loss. Hence, the deep network complexity is adapted by proposed novel restrained learning, which introduces dynamic estimation of the network depth. The efficacy of Blend-Res2Net is demonstrated by a series of ablation experiments over publicly available benchmark time series archive- UCR. We further establish the superior performance of Blend-Res2Net over baselines and state-of-the-art algorithms including 1-NN-DTW, HIVE-COTE, ResNet, InceptionTime, ROCKET, DMS-CNN, TS-Chief. |
