Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Estimating Uncertainty Online Against an Adversary
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use bucketed online recalibration of the baseline forecasts | Baseline forecaster F · Bernoulli sequence | calibration error | Helped |
| Use bucketed online recalibration instead of RegMin alone | RegMin · Bernoulli sequence | ℓ2 loss | Helped |
| Use bucketed online recalibration on noisy baseline forecasts | Baseline forecaster F · Adversarially chosen outcomes | calibration error | Helped |
| Use bucketed online recalibration | Question-answering system · Free917 | calibration | Helped |
| Use bucketed online recalibration | Online ℓ1-regularized linear SVM · Type 1 diabetes genomic diagnosis | calibration error | Helped |
| Use bucketed online recalibration | Online ℓ1-regularized linear SVM · Type 1 diabetes genomic diagnosis | accuracy (ℓ2 loss): 0.01 | Mixed |
| Recalibrate with a proper loss rather than a non-proper loss | Online recalibration · Adversarial sequences | accuracy | Helped |
| Use recalibration instead of predicting the calibrated constant 1/3 under ℓ1 loss | Online recalibration · Repeating 001 sequence | ℓ1 loss: 1/3 < 0.44 | Hurt |
Closest papers by meaning
- Online Algorithms with Uncertainty-Quantified Predictions (ICML 2024 · similarity 0.87)
- Towards Certification of Uncertainty Calibration under Adversarial Attacks (ICLR 2025 · similarity 0.86)
- Probably Approximately Global Robustness Certification (ICML 2025 · similarity 0.85)
- Experts Don't Cheat: Learning What You Don't Know By Predicting Pairs (ICML 2024 · similarity 0.85)
- Uncertainty Quantification with the Empirical Neural Tangent Kernel (NeurIPS 2025 · similarity 0.85)
- Uncertainty for Active Learning on Graphs (ICML 2024 · 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.