AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Rapid Network Adaptation: Learning to Adapt Neural Networks Using Test-Time Feedback

Teresa Yeo, Oğuzhan Fatih Kar, Zahra Sodagar, Amir Zamir

ICCV 2023 · Label-efficient and robust learning › Test-time adaptation · approach: Realistic and stable test-time adaptation

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
replace test-time optimization with learned amortized adaptation (RNA)depth adaptation · Taskonomy depthrelative runtime: 14.85 -> 1.56Improved
use RNA instead of TTO (episodic)depth Taskonomy-CC · Taskonomy-CCl1 error x100: 2.99 -> 1.68Improved
RNA vs online TTO on Hypersim cross-dataset shiftdepth · Hypersiml1 error x100: TTO online 14.17 vs RNA 16.17Hurt
train h jointly with f instead of frozen f (ablation)RNA · Taskonomy-CC depthl1 error x100: 1.68 -> 1.50Improved
use FiLM-f controller instead of FiLM-x or HyperNetwork-x (ablation)RNA · Taskonomy-CC depthl1 error: 0.039 / 0.041 -> 0.024Improved
feed concatenated RGB and sparse supervision to a single modelmulti-modal baseline · Taskonomy-CC deptherror vs RNA: 42.5% worse than RNAHurt
use coarse WordNet labels with RNAResNet50 · ImageNet cleanerror (%): TTO 24.72 -> RNA 16.72Improved
use noisy DINO k-NN coarse labels with RNAResNet50 · ImageNet-Cerror (%): TENT 46.19 vs RNA 54.86Hurt
control for parameter count across methods (ablation)RNA · Taskonomy-CC depthl1 error x100: TTO 0.019 -> RNA 0.018Improved
apply RNA to DPT and ConvNext main networks (ablation)DPT, ConvNext · Taskonomy-CC / ImageNet-Cerror: DPT depth 2.61 -> 1.56; ConvNext 41.44 -> 38.04Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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