AI Paper Atlas

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

Training for the Future: A Simple Gradient Interpolation Loss to Generalize Along Time

Anshul Nasery, Soumyadeep Thakur, Vihari Piratla, Abir De, Sunita Sarawagi

NeurIPS 2021 · Label-efficient and robust learning › Test-time adaptation · approach: Characterizing out-of-distribution failures · 6 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
Use GI instead of DPerturbTemporal generalization models · 2-Moonsmisclassification error (%): 5.6 ± 2.6 -> 3.5 ± 1.4Improved
Use GI instead of Generative ModelTemporal generalization models · Rot-MNISTmisclassification error (%): 13.5 ± 3.2 -> 7.7 ± 1.3Improved
Use GI instead of DPerturbTemporal generalization models · M5-HobMAE: 0.28 ± 0.09 -> 0.09 ± 0.03Improved
Use Deep ODE instead of GITemporal generalization models · 2-Moonsmisclassification error (%): 3.5 ± 1.4 -> 2.2 ± 0.8Improved
Use GI instead of FTPLOnline-learning comparison · Rot-MNISTerror: 11.5 ± 3.1 -> 7.7 ± 1.3Improved
Use GI instead of FTRLOnline-learning comparison · Elec2error: 20.7 ± 0.8 -> 16.9 ± 0.7Improved
Use GI instead of FTPLOnline-learning comparison · Shuttleerror: 0.75 ± 0.12 -> 0.21 ± 0.05Improved
Use GI instead of FTRLOnline-learning comparison · Houseerror: 10.3 ± 0.3 -> 9.6 ± 0.02Improved
Use GI instead of CIDATemporal generalization training · Rot-MNISTtraining time (seconds): 257.3 -> 207.1Improved
Use GI instead of ERMTemporal generalization training · 2-Moonstraining time (seconds): 6.2 -> 7.1Hurt
Use one shared adversarial δ per minibatch rather than a separate δ per example (ablation)GI fine-tuning · Reported across datasetsperformanceImproved

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