AI Paper Atlas

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

Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation

Muhammad Ghifary, W. Bastiaan Kleijn, Mengjie Zhang, David Balduzzi, Wen Li

ECCV 2016 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adversarial domain adaptation · 962 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
reconstruct unlabeled target images only rather than unlabeled source images (DRCN vs DRCN_s) (ablation)DRCN · SVHN->MNISTaccuracy (%): 73.66 -> 81.97Improved
reconstruct unlabeled target images only rather than source+target (DRCN vs DRCN_st) (ablation)DRCN · SVHN->MNISTaccuracy (%): 75.88 -> 81.97Improved
reconstruct target only vs source only (ablation)DRCN · USPS->MNISTaccuracy (%): 65.96 -> 73.67Improved
reconstruct target only vs source+target (ablation)DRCN · USPS->MNISTaccuracy (%): 68.64 -> 73.67Improved
reconstruct target only vs source only (ablation)DRCN · MNIST->SVHNaccuracy (%): 34.29 -> 40.05Improved
reconstruct target only vs source+target (ablation)DRCN · STL->CIFARaccuracy (%): 55.26 -> 58.86Improved
reconstruct target only vs source+target (ablation)DRCN · MNIST->USPSaccuracy (%): 91.15 -> 91.80No clear effect
reconstruct target only vs source only (ablation)DRCN · CIFAR->STLaccuracy (%): 63.02 -> 66.37Improved
replace adversarial domain-confusion training (ReverseGrad) with joint reconstruction training (DRCN)DRCN · SVHN->MNISTaccuracy (%): 73.91 -> 81.97Improved
replace ReverseGrad with DRCNDRCN · MNIST->SVHNaccuracy (%): 35.67 -> 40.05Improved
replace ReverseGrad with DRCNDRCN · USPS->MNISTaccuracy (%): 74.01 -> 73.67Hurt
joint reconstruction-classification training instead of greedy pretraining then finetuning (SCAE_t)DRCN · SVHN->MNISTaccuracy (%): 65.57 -> 81.97Improved

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