AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

On the Dynamics Under the Unhinged Loss and Beyond

Xiong Zhou, Xianming Liu, Hanzhang Wang, Deming Zhai, Junjun Jiang, Xiangyang Ji

JMLR 2023 · Optimization › Implicit bias and generalization · approach: Neural tangent kernel analysis · 0 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 explicit feature regularization (lambda=5e-5) (ablation)CE · Long-tailed CIFAR-10, ratio 100accuracy: 67.81 -> 69.74Improved
add explicit feature regularization (lambda=5e-5) (ablation)CE · Long-tailed CIFAR-100, ratio 20accuracy: 42.96 -> 54.72Improved
add explicit feature regularization (lambda=1e-5) (ablation)CE · Long-tailed CIFAR-100, ratio 100accuracy: 33.37 -> 36.61Improved
add explicit feature regularization (lambda=1e-6) (ablation)unhinged loss · Long-tailed CIFAR-10, ratio 100accuracy: 72.60 -> 73.71Improved
add explicit feature regularization (lambda=1e-7) (ablation)unhinged loss · Long-tailed CIFAR-10, ratio 100accuracy: 72.60 -> 71.85Hurt
add explicit feature regularization (lambda=5e-6) (ablation)unhinged loss · Long-tailed CIFAR-10, ratio 100accuracy: 72.60 -> 71.13Hurt
add explicit feature regularization (lambda=5e-7) (ablation)unhinged loss · Long-tailed CIFAR-100, ratio 100accuracy: 39.20 -> 41.85Improved
add explicit feature regularization (lambda=5e-6 or 1e-5) (ablation)CE · balanced CIFAR-10, ratio 1accuracy: 95.28 -> 95.27No clear effect
add explicit feature regularization (lambda=1e-5) (ablation)CE · balanced CIFAR-100, ratio 1accuracy: 75.42 -> 77.34Improved
add explicit feature regularization (lambda=5e-6) (ablation)unhinged loss · balanced CIFAR-100, ratio 1accuracy: 75.37 -> 77.20Improved
add explicit feature regularization (EFL) (ablation)CE · OOD detection CIFAR-10 vs SVHN, energy-basedFPR95: 43.04 -> 27.87Improved
add explicit feature regularization (EFL) (ablation)unhinged loss · OOD detection CIFAR-10 vs SVHN, energy-basedFPR95: 53.33 -> 27.24Improved

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