Research map / Learning theory and algorithms / Prediction and decision losses
TESSAR: Geometry-Aware Active Regression via Dynamic Voronoi Tessellation
Seong Jin Cho, Gwangsu Kim, Junghyun Lee, Hee Suk Yoon, Joshua Tian Jin Tee, Chang D. Yoo
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add DIST to VLDM (ablation) | TESSAR · Protein, Road, Stock | RMSE | Improved |
| add BIN density term to VLDM+DIST (ablation) | TESSAR · Protein, Road, Stock | RMSE | Improved |
| use VLDM instead of margin-based proxy (ablation) | TESSAR · Protein, Road, Stock | RMSE | Improved |
| use dynamic Voronoi updates instead of static (ablation) | TESSAR · Protein, Road, Stock | RMSE | Improved |
| use dynamic Voronoi updates for VLDM alone (ablation) | VLDM · Protein, Road, Stock | RMSE | Improved |
| vary number of perturbations N in {10, 50, 100, 500, 1000} (ablation) | TESSAR · Protein, Road, Stock | RMSE | No clear effect |
| use TESSAR instead of LCMD | active regression · 14 tabular datasets (performance profile) | RMSE: R(0) 41% vs 29% | Improved |
| use TESSAR instead of LCMD | active regression · Road | RMSE: 0.0001 vs 0.0036 | Improved |
| use TESSAR instead of LCMD | active regression · WEC Sydney | RMSE: -0.0001 vs 0.0010 | Improved |
| use TESSAR instead of LCMD | active regression · Query | RMSE: -0.0047 vs -0.0054 | Hurt |
| use TESSAR instead of BAIT | active regression · Friedman | RMSE: -0.0030 vs -0.0049 | Hurt |
Closest papers by meaning
- A Framework and Benchmark for Deep Batch Active Learning for Regression (JMLR 2023 · similarity 0.86)
- STARS: Spatial-Temporal Active Re-sampling for Label-Efficient Learning from Noisy Annotations (AAAI 2023 · similarity 0.86)
- Margin-based sampling in high dimensions: When being active is less efficient than staying passive (ICML 2023 · similarity 0.86)
- Winner-takes-all learners are geometry-aware conditional density estimators (ICML 2024 · similarity 0.84)
- Boundary Matters: A Bi-Level Active Finetuning Method (NeurIPS 2024 · similarity 0.84)
- TTVD: Towards a Geometric Framework for Test-Time Adaptation Based on Voronoi Diagram (ICLR 2025 · similarity 0.84)
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.
