Research map / Label-efficient and robust learning / Recognition and long-tail
Enhancing Deep Batch Active Learning for Regression with Imperfect Data Guided Selection
Yinjie Min, Furong Xu, Xinyao Li, Changliang Zou, Yongdao Zhou
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 active learning method for deep regression that combines gradient-based model sensitivity with predictive uncertainty, guided by imperfect data to estimate uncertainty without true labels.
Problem
Deep batch active learning for regression struggles because predictive uncertainty cannot be estimated without true labels, which reduces the informativeness of selected samples, even though model sensitivity can be measured unsupervised via parameter gradients.
Approach
Combines parameter-gradient sensitivity with an uncertainty estimate obtained by guiding selection with imperfect data, so that predictive uncertainty can be approximated for regression without access to true labels.
Closest papers by meaning
- Deep Active Learning with Noise Stability (AAAI 2024 · similarity 0.86)
- Transductive Active Learning: Theory and Applications (NeurIPS 2024 · similarity 0.86)
- Robust Sampling for Active Statistical Inference (NeurIPS 2025 · similarity 0.86)
- Prediction-Powered Active Testing (NeurIPS 2026 · similarity 0.85)
- Rethinking Epistemic and Aleatoric Uncertainty for Active Open-Set Annotation: An Energy-Based Approach (CVPR 2025 · similarity 0.85)
- Debiased Active Learning with Variational Gradient Rectifier (AAAI 2025 · similarity 0.85)
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.
