Skip to content

Commit 88614f1

Browse files
committed
change
1 parent 65696c0 commit 88614f1

File tree

1 file changed

+1
-1
lines changed

1 file changed

+1
-1
lines changed

README.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -20,7 +20,7 @@ Tianyi Liu*, Zihao Xu*, Hao He, Guang-Yuan Hao, Guang-He Lee, Hao Wang<br>
2020
For classical domain adaptation methods such as DANN, they enforce uniform alignment to boost the generalization ability of models. However, recent studies have shown that, such uniform alignment can harm domain adaptation performance. To deal with this problem, we incorporate domain taxonomy into domain adaptation process. With domain taxonomy, we can **break** the uniform alignment in domain adaptation. The equilibrium recovers the classic adversarial domain adaptation’s solution if given a non-informative domain taxonomy (e.g., a flat taxonomy where all leaf nodes connect to the root node) while yielding non-trivial results with other taxonomies. See Figure 1 for an example.
2121

2222
<p align="center">
23-
<img src="fig/example_taxonomy.png" alt="" data-canonical-src="fig/example_taxonomy.png" width="93%"/>
23+
<img src="fig/example_taxonomy.png" alt="" data-canonical-src="fig/example_taxonomy.png" width="80%"/>
2424
</p>
2525
<p>
2626
<em>Figure 1. An example of using domain taxonomy to break the uniform alignment. For our model, the middle representation for basset and Beagle should be more similar than the one for basset and tabby.</em>

0 commit comments

Comments
 (0)