CaltechAUTHORS
  A Caltech Library Service

View-Adaptive Weighted Deep Transfer Learning for Distributed Time-Series Classification

Das Bhattacharjee, Sreyasee and Tolone, William J. and Mahabal, Ashish and Elshambakey, Mohammed and Cho, Isaac and Djorgovski, George (2019) View-Adaptive Weighted Deep Transfer Learning for Distributed Time-Series Classification. In: 2019 IEEE 43rd Annual Computer Software and Applications Conference (COMPSAC). IEEE , Piscataway, NJ, pp. 373-381. ISBN 9781728126074. http://resolver.caltech.edu/CaltechAUTHORS:20190711-134821741

Full text is not posted in this repository. Consult Related URLs below.

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20190711-134821741

Abstract

In this paper, we propose an effective, multi-view, deep, transfer learning framework for multivariate time-series data. Though widely used for tasks such as computer vision, the application of transfer learning to time-series classification problems (e.g., classification of light curves) is underexplored. The proposed framework makes several important contributions to facilitate knowledge sharing, while simultaneously ensuring an effective solution for domain specific fine-level categorizations. First, in contrast to the traditional approaches, the proposed framework describes pairwise view similarity by identifying a smaller subset of source-view samples that closely resemble the target data patterns. Second, by means of two-phase learning, a generic baseline model is learned on a larger source data collection and later fine-tuned on a smaller target data collection, precisely approximating the target data patterns. Third, an effective view-adaptive timestamp weighting scheme evaluates the relative importance of each timestamp in a more data-driven manner, which enables a more flexible yet discriminative feature representation scheme in the presence of evolving data characteristics. As shown by experiments, compared to the existing approaches, our proposed deep transfer learning framework improves classification performance by around 2-3% in the UCI multi-view activity recognition dataset, while also showing a robust, generalized representation capacity in classifying several large-scale multi-view light curve collections.


Item Type:Book Section
Related URLs:
URLURL TypeDescription
https://doi.org/10.1109/compsac.2019.00061DOIArticle
ORCID:
AuthorORCID
Mahabal, Ashish0000-0003-2242-0244
Djorgovski, George0000-0002-0603-3087
Additional Information:© 2019 IEEE. Funding for this research was provided by the National Science Foundations (NSF) Data Infrastructure Building Blocks (DIBBs) Progam under award #1640818.
Funders:
Funding AgencyGrant Number
NSFOAC-1640818
Subject Keywords:LSTM, RNN, Multi-view Classification, Transfer Learning, Distributed Time-Series Analysis, Deep Learning, Minimum Description Length
Record Number:CaltechAUTHORS:20190711-134821741
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20190711-134821741
Official Citation:S. Das Bhattacharjee, W. J. Tolone, A. Mahabal, M. Elshambakey, I. Cho and G. Djorgovski, "View-Adaptive Weighted Deep Transfer Learning for Distributed Time-Series Classification," 2019 IEEE 43rd Annual Computer Software and Applications Conference (COMPSAC), Milwaukee, WI, USA, 2019, pp. 373-381. doi: 10.1109/COMPSAC.2019.00061
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:97063
Collection:CaltechAUTHORS
Deposited By: Tony Diaz
Deposited On:11 Jul 2019 22:35
Last Modified:11 Jul 2019 22:35

Repository Staff Only: item control page