Research map / Bandits and online learning / Online convex optimization and games
Random Permutation Online Isotonic Regression
Wojciech Kotlowski, Wouter M. Koolen, Alan Malek
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 |
|---|---|---|---|
| Use the online-to-batch conversion of the optimal fixed-design algorithm (random permutation learner from Eq. 3) | online isotonic regression in random permutation model · theoretical regret bound | excess leave-one-out loss / regret | Improved |
| Forward algorithms (any label estimate) analyzed in random permutation model | forward algorithms · theoretical regret bound | excess leave-one-out loss | No clear effect |
| Forward algorithms analyzed in well-specified (i.i.d. noise around isotonic means) setting | forward algorithms · well-specified case | excess leave-one-out loss | Improved |
| Use IR-Int (interpolation) forward algorithm on adversarial alternating-segment label sequence | IR-Int forward algorithm · hard label sequence (segments with proportion k/n of ones) | excess leave-one-out loss | Hurt |
| Heavy-γ with weight c=Θ(t^{1/3}) on the current instance (conjectured) | forward algorithm with heavy weighting · theoretical conjecture | excess leave-one-out loss | No clear effect |
| Log-IVAP forward algorithm under entropic loss instead of squared loss | log-IVAP · entropic loss | excess leave-one-out loss | No clear effect |
| Plugging in the IR-Int label estimate into any weighted forward scheme (Heavy-type) with any weight | Heavy-style forward algorithms · theoretical | excess leave-one-out loss | Hurt |
Closest papers by meaning
- From Average Sensitivity to Small-Loss Regret Bounds under Random-Order Model (NeurIPS 2026 · similarity 0.87)
- Online Learning in the Random-Order Model (ICML 2025 · similarity 0.86)
- Efficient Algorithms for Non-convex Isotonic Regression through Submodular Optimization (NeurIPS 2018 · similarity 0.86)
- Isotonic Modeling with Non-Differentiable Loss Functions with Application to Lasso Regularization (TPAMI 2016 · similarity 0.86)
- Are Greedy Task Orderings Better Than Random in Continual Linear Regression? (NeurIPS 2025 · similarity 0.84)
- A Simple and Optimal Approach for Universal Online Learning with Gradient Variations (NeurIPS 2024 · 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.
