AI Paper Atlas

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

Label Efficient Learning of Transferable Representations acrosss Domains and Tasks

Zelun Luo, Yuliang Zou, Judy Hoffman, Li F Fei-Fei

NeurIPS 2017 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Statistical transfer learning

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 the full multi-layer domain and semantic transfer objectivesfine-tuning from SVHN to MNIST · MNIST 5-9 test, k=2accuracy: 0.612 ± 0.020 -> 0.917 ± 0.007Improved
add multi-layer adversarial domain adaptation without the full semantic transfer objective (ablation)fine-tuning from SVHN to MNIST · MNIST 5-9 test, k=2accuracy: 0.612 ± 0.020 -> 0.702 ± 0.020Improved
add semantic transfer to fine-tuning plus adversarial adaptation (ablation)SVHN-to-MNIST transfer · MNIST 5-9 test, k=2accuracy: 0.702 ± 0.020 -> 0.917 ± 0.007Improved
use the full method rather than fine-tuningImageNet-to-UCF-101 transfer · UCF-101, k=3per-video accuracy: 0.406±0.015 -> 0.467±0.007Improved
use the full method rather than fine-tuningImageNet-to-UCF-101 transfer · UCF-101, k=5per-video accuracy: 0.523±0.010 -> 0.545±0.014Improved
use the full method rather than fine-tuningImageNet-to-UCF-101 transfer · UCF-101, k=10per-video accuracy: 0.568±0.042 -> 0.620±0.005Improved
use the full method rather than fine-tuningImageNet-to-UCF-101 transfer · UCF-101, k=5per-frame accuracy: 0.486±0.012 -> 0.459±0.013Hurt
use the full method rather than fine-tuningImageNet-to-UCF-101 transfer · UCF-101, k=10per-frame accuracy: 0.529±0.039 -> 0.523±0.002No clear effect
replace the competing adversarial adaptation approach with the proposed multi-layer discriminator (ablation)unsupervised SVHN-to-MNIST adaptation · MNIST testaccuracy: 0.760 ± 0.018 -> 0.810 ± 0.003Improved
use domain adversarial learning rather than source-only trainingunsupervised SVHN-to-MNIST adaptation · MNIST testaccuracy: 0.601 ± 0.011 -> 0.810 ± 0.003Improved
replace the full method with target-only training from scratchSVHN-to-MNIST transfer · MNIST 5-9 test, k=2accuracy: 0.917 ± 0.007 -> 0.642 ± 0.026Hurt

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