Research map / Graph learning / Graph algorithms and clustering
Adaptive Annealing for Robust Averaging
Sidhartha Chitturi, Venu Madhav Govindu
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 |
|---|---|---|---|
| replace fixed-factor annealing with adaptive Hessian-guided annealing (GNCp) | GNC for vector averaging · Vector averaging synthetic, outlier fraction 0.5 | mean error: 13.99 -> 1.37 vs GNC (delta down) | Improved |
| use adaptive annealing instead of large fixed delta (delta=5) | GNC for vector averaging · Vector averaging, outlier fraction 0.5 | mean error: 86.26 -> 1.37 | Improved |
| adaptive annealing vs small fixed delta | GNC for vector averaging · Vector averaging, outlier fraction 0.5 | time (s): 15.37 -> 4.9 | Improved |
| large fixed delta annealing vs adaptive annealing | GNC for vector averaging · Vector averaging, outlier fraction 0.5 | time (s): 3.56 vs 4.9 (fast but poor accuracy) | Hurt |
| adaptive annealing at low outlier fraction | GNC for vector averaging · Vector averaging, outlier fraction 0.1 | mean error: 0.41 vs 0.41 | No clear effect |
| adaptive annealing on real indoor data | Augmented ICL-NUIM · Augmented ICL-NUIM | mean error: 1.93 vs 1.93 (lr1) | No clear effect |
| adaptive annealing on outlier filtering | GNC for vector averaging · Vector averaging, outlier fraction 0.5 | outliers remaining (%): 55.2 -> 0 | Improved |
| GNCp-LUD with Geman-McClure vs original LUD on real 1DSfM data | Translation averaging · 1DSfM datasets | mean/median error: superior in 9 (mean) and 11 (median) of 15 datasets | Mixed |
| GNCp-LUD on Trafalgar | Translation averaging · 1DSfM Trafalgar | mean error: 8.4 -> 10.4 vs LUD | Hurt |
Closest papers by meaning
- Adaptive Annealing for Robust Geometric Estimation (CVPR 2023 · similarity 0.90)
- SAC-GNC: SAmple Consensus for adaptive Graduated Non-Convexity (ICCV 2025 · similarity 0.90)
- Robust Averaging using Adaptive Annealing (IJCV 2026 · similarity 0.85)
- Convergence guarantees for a class of non-convex and non-smooth optimization problems (ICML 2018 · similarity 0.85)
- An Online Adaptive Sampling Algorithm for Stochastic Difference-of-convex Optimization with Time-varying Distributions (ICML 2025 · similarity 0.83)
- Transfer Learning on Edge Connecting Probability Estimation Under Graphon Model (NeurIPS 2025 · similarity 0.83)
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.
