AI Paper Atlas

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

Say No to Freeloader: Protecting Intellectual Property of Your Deep Model

Lianyu Wang, Meng Wang, Huazhu Fu, Daoqiang Zhang

TPAMI 2024 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Source-free domain adaptation · 5 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
add L_t loss to the CUPI-Domain generators (ablation)SL + CUPI generators · MNIST -> SVHNunauthorized domain accuracy: 11.4 -> 9.4Improved
add L_i loss to the CUPI-Domain generators (ablation)SL + CUPI generators · MNIST -> SVHNunauthorized domain accuracy: 9.4 -> 9.4No clear effect
add L_i loss to the CUPI-Domain generators (ablation)SL + CUPI generators · CIFAR10 -> STL10unauthorized domain accuracy: 11.4 -> 11.2Improved
add DIMB with L_Stl and L_Dis (ablation)CUPI-Domain generators · MNIST -> SVHNunauthorized domain accuracy: 9.4 -> 6.1Improved
add L_Stl loss (ablation)CUPI-Domain with DIMB · MNIST -> SVHNunauthorized domain accuracy: 6.3 -> 6.1Improved
add L_Dis loss (ablation)CUPI-Domain with DIMB · VisDA-T -> VisDA-Vunauthorized domain accuracy: 9.6 -> 7.5Improved
add CUPI-Domain generators to supervised learning (ablation)SL · MNIST -> SVHNunauthorized domain accuracy: 38.2 -> 9.4Improved
use CUPI-Domain instead of CUTI-Domain, target-specifieddigit datasets · Digit datasetsauthorized domain accuracy drop: 0.13 -> 0.02Improved
use CUPI-Domain instead of NTL, target-specifieddigit datasets · Digit datasetsunauthorized domain accuracy drop: 46.48 -> 56.83Improved
use CUPI-Domain instead of CUTI-Domain, target-freedigit datasets · Digit datasetsunauthorized domain accuracy drop: 53.47 -> 55.10Improved
use CUPI-Domain instead of NTL in applicability authorization scheme 1digits, CIFAR10&STL10, VisDA · Scheme-1accuracy drop: 81.03 -> 88.57Improved
use CUPI-Domain instead of CUTI in applicability authorization scheme 2digits, CIFAR10&STL10, VisDA · Scheme-2accuracy drop: 85.85 -> 85.50No clear effect

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