Abstract | ||
---|---|---|
•A novel metric-based regularization for domain-invariant training of neural networks.•Some theoretical properties of the proposed moment-based metric are analyzed.•Tested on benchmark datasets for sentiment analysis, object recognition and digit recognition. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1016/j.ins.2019.01.025 | Information Sciences |
Keywords | DocType | Volume |
Transfer learning,Domain adaptation,Neural networks,Moment distance,Integral probability metric | Journal | 483 |
ISSN | Citations | PageRank |
0020-0255 | 5 | 0.43 |
References | Authors | |
36 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Werner Zellinger | 1 | 32 | 4.27 |
Bernhard Moser | 2 | 36 | 9.78 |
Thomas Grubinger | 3 | 9 | 1.52 |
Edwin Lughofer | 4 | 1940 | 99.72 |
T Natschläger | 5 | 1199 | 102.98 |
Susanne Saminger-Platz | 6 | 76 | 10.94 |