AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Open Set Domain Adaptation by Backpropagation

Kuniaki Saito, Shohei Yamamoto, Yoshitaka Ushiku, Tatsuya Harada

ECCV 2018 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adversarial domain adaptation · 607 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use adversarial unknown-class boundary (t=0.5) instead of OSVM on source-trained featuresopen set domain adaptation · Office 10 classes (AlexNet)OS accuracy (%): 40.6 -> 80.4Improved
use adversarial unknown-class boundary instead of OSVMopen set domain adaptation · Office 10 classes (VGGNet)OS accuracy (%): 65.3 -> 88.0Improved
add MMD distribution matching before OSVMOSVM · Office 10 classes (AlexNet)OS accuracy (%): 40.6 -> 34.5Hurt
add domain-classifier (BP) distribution matching before OSVMOSVM · Office 10 classes (AlexNet)OS accuracy (%): 40.6 -> 29.5Hurt
add MMD distribution matching before OSVMOSVM · Office 10 classes (VGGNet)OS accuracy (%): 65.3 -> 66.3No clear effect
use adversarial unknown-class boundary instead of OSVMopen set domain adaptation · Office 20 classes (VGGNet)OS accuracy (%): 61.8 -> 74.7Improved
use adversarial unknown-class boundary instead of OSVMopen set domain adaptation · VisDA (AlexNet)Avg accuracy over classes (%): 37.8 -> 58.5Improved
use adversarial unknown-class boundary instead of OSVMopen set domain adaptation · VisDA (VGGNet)Avg accuracy over classes (%): 52.5 -> 65.2Improved
use adversarial unknown-class boundary instead of BP + OSVMopen set domain adaptation · DigitsUNK accuracy (%): 15.3 -> 85.9Improved
use adversarial unknown-class boundary instead of OSVMopen set domain adaptation · DigitsOS accuracy (%): 59.1 -> 82.4Improved
increase unknown-class target probability t toward 1 (ablation)adversarial open set DA · Office DSLR to AmazonOS and OS* accuracy: decrease as t increasesHurt
add OSVM on top of the learned featuresOurs · Semi-supervised open set recognition (SVHN)UNK accuracy (%): 19.2 -> 93.1Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy