Abstract | ||
---|---|---|
In domain adaptation (DA), label-induced losses generally occupy a dominant position and most previous models regard hard or soft labels as their inputs. However, these two types of labels may mislead the modeling process of label-induced losses since hard label is sensitive to a wrongly-predicted sample while soft label may introduce label noise, thus they may cause negative transfer. To relieve this problem, we propose a novel label learning approach namely confidence regularized label propagation (CRLP) that regularizes the confidence of predicted soft labels with constraints of F-norm or L
<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">21</sub>
-norm. It is validated that maximizing either one of these two constraints equals to minimizing entropy loss. Specially, we illustrate that L
<sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">21</sub>
-norm is more suitable for DA than F-norm when the dataset contain a large number of categories. Then, we leverage the regularized soft labels produced by CRLP to reformulate some popular label-induced losses that consider feature transferability and discriminability such as class-wise maximum mean discrepancy, intra-class compactness and inter-class dispersion in a probability manner to present a novel DA method (
<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e.</i>
, CRLP-DA). Comprehensive analysis and experiments on four cross-domain object recognition datasets verify that the proposed CRLP-DA outperforms some state-of-the-art methods, especially 59.5% for Office10+Caltech10 dataset with SURF features. For others to better reproduce, our preliminary Matlab code will be available at
<uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/WWLoveTransfer/CRLP-DA/</uri>
. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/TCSVT.2021.3104835 | IEEE Transactions on Circuits and Systems for Video Technology |
Keywords | DocType | Volume |
Domain adaptation,hard label,soft label,label propagation,entropy loss,F/L₂₁-norm,label-induced losses | Journal | 32 |
Issue | ISSN | Citations |
6 | 1051-8215 | 0 |
PageRank | References | Authors |
0.34 | 33 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wei Wang | 1 | 0 | 0.34 |
Baopu Li | 2 | 348 | 30.88 |
Mengzhu Wang | 3 | 2 | 3.44 |
Feiping Nie | 4 | 7061 | 309.42 |
zhihui wang | 5 | 39 | 12.86 |
Haojie Li | 6 | 1427 | 65.70 |