AI Paper Atlas

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

Beyond Sharing Weights for Deep Domain Adaptation

Artem Rozantsev, Mathieu Salzmann, Pascal Fua

TPAMI 2018 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adversarial domain adaptation · 488 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 MMD loss to the two-stream loss (ablation)Ours (Ls+Lt+Lw) · UAV-200 (full, supervised DA)AP: 0.673 -> 0.732Improved
add weight-linking term Lw to Ls+Lt+LMMD (ablation)Ours · UAV-200 (full, supervised DA)AP: 0.711 -> 0.732Improved
use non-shared weights with regularizer instead of shared weights (DDC)two-stream CNN · UAV-200 (full, supervised DA)AP: 0.664 -> 0.732Improved
add synthetic data pre-training and fine-tuning on real data (vs real only)CNN · UAV-200 (full, supervised DA)AP: 0.575 -> 0.732Improved
naively train on real and synthetic together (Ls+Lt)CNN · UAV-200 (full, supervised DA)AP: 0.575 -> 0.569No clear effect
replace non-shared weights by DDC shared weightstwo-stream CNN · UAV-200 (small, supervised DA)accuracy: 0.92 -> 0.89Hurt
drop weight-linking terms while still minimizing MMD (ablation)Ours · UAV-200 (full, supervised DA)AP: 0.732 -> 0.711Hurt
choose layer sharing by lowest MMD instead of best validation architectureUAV architecture selection · UAV-200 (small, supervised DA)test accuracy: 0.757 -> 0.732No clear effect
allow non-shared weights in last two layers vs shared (GRL)two-stream AlexNet · Office A->Waccuracy: 0.730 -> 0.760Improved
use non-shared weights instead of shared (DDC)two-stream CNN · MNIST-USPS (average)accuracy: 0.554 -> 0.640Improved
use non-shared weights instead of shared (DDC)two-stream regressor · Facial landmark pose estimationaverage PCP score (%): 60.3 -> 62.8Improved
add more real positive samples (200 to 5000) with 32800 syntheticOurs · UAV-200 (full, supervised DA)APImproved

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