AI Paper Atlas

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

Learning model uncertainty as variance-minimizing instance weights

Nishant Jain, Karthikeyan Shanmugam, Pradeep Shenoy

ICLR 2024 · Label-efficient and robust learning › Test-time adaptation · approach: Extrapolation and single-sample adaptation

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes an instance-conditional reweighting method that learns predictive uncertainty via an auxiliary network producing variance-minimizing instance weights, unifying train-time robustness interventions and test-time selective classification.

Problem

Capturing a model's predictive uncertainty (self-awareness of its accuracy on an input) for building robust models via training interventions and for test-time applications such as selective classification.

Approach

An auxiliary network predicts instance-conditional weights that reweight training examples so as to minimize variance, yielding a learned uncertainty estimate usable both during training and at test time.

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