-
https://arxiv.org/pdf/1607.03516.pdf
In this paper, we propose a novel unsupervised domain adaptation algorithm based on deep learning for visual object recognition. Specifically, we design a new m…
leo-p updated
7 years ago
-
## 一言でいうと
CycleGANにおける再構成制約を緩和した研究。CycleGANではS=>T=>S'と変換した際S/S'の同一性を見るが、これは「同ドメイン(スタイル)」かつ「同コンテンツ」である必要があり、データが少ない場合学習が難しかった。そこで「同コンテンツ」であればOKにする緩和を行なっている
![image](https://user-images.githubuser…
-
https://arxiv.org/abs/2006.00632
-
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.110.2062&rep=rep1&type=pdf
-
When using `SetFit` for classification in a more technical domain, I could imagine the generically-trained `SBERT` models may produce poor sentence embeddings if the domain is not represented well eno…
-
The paper by [Bateson et al.](https://arxiv.org/pdf/1908.02996.pdf) seems relevant for this problem.
The task is simpler that in the paper of Bateson et al.: it only consists of a semantic labeling…
-
- [ ] Read and summary [Agent-Specific Deontic Modality Detection in Legal Language](https://aclanthology.org/2022.emnlp-main.795.pdf).
- [ ] Read and summary [What to Read in a Contract? Party-Speci…
-
# Overview
## Universal Domain Adaptation through Self Supervision
https://arxiv.org/abs/2002.07953
![image](https://user-images.githubusercontent.com/6381645/85234560-68433c00-b40e-11ea-999…
-
Can we use test image when training the domain adaptation model?
-
Hi, thanks for your wonderful work. From the paper, the batch normalization statistics updates for combined source and target images, and keep fixed when computing source images for the second time. H…