Research map / Bandits and online learning / Online convex optimization and games
Parameter-free non-stationary online learning
Achieve universal, parameter-free guarantees in changing environments through meta-algorithm ensembles.
45 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Online convex optimization and games, keyword cluster “convex optimization · online convex · functions”
State History Tried Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
12 papers in 2023–24 and 5 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 5.2% to 1.8%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · MetaGrad: Multiple Learning Rates in Online Learning (NeurIPS 2016 · 20 citations)
- 2017 · Projection-free Distributed Online Learning in Networks (ICML 2017 · 48 citations)
- 2018 · Dynamic Regret of Strongly Adaptive Methods (ICML 2018 · 33 citations)
- 2019 · Adaptive Scale-Invariant Online Algorithms for Learning Linear Models (ICML 2019 · 12 citations)
- 2020 · Online mirror descent and dual averaging: keeping pace in the dynamic case (ICML 2020 · 18 citations)
- 2021 · Projection-free Online Learning in Dynamic Environments (AAAI 2021 · 8 citations)
- 2022 · PDE-Based Optimal Strategy for Unconstrained Online Learning (ICML 2022 · 0 citations)
- 2023 · Meta-Learning Adversarial Bandit Algorithms (NeurIPS 2023 · 2 citations)
- 2024 · Non-stationary Projection-Free Online Learning with Dynamic and Adaptive Regret Guarantees (AAAI 2024 · 4 citations)
- 2025 · Label Shift Meets Online Learning: Ensuring Consistent Adaptation with Universal Dynamic Regret (CVPR 2025 · 0 citations)
- 2026 · Non-Stationary Online Structured Prediction with Surrogate Losses (ICML 2026 · 0 citations)
Earliest papers here
- MetaGrad: Multiple Learning Rates in Online Learning (NeurIPS 2016 · 20 citations)
- Combining Adversarial Guarantees and Stochastic Fast Rates in Online Learning (NeurIPS 2016 · 19 citations)
- Following the Leader and Fast Rates in Linear Prediction: Curved Constraint Sets and Other Regularities (NeurIPS 2016 · 11 citations)
What has been tried
35 papers here report 204 measured changes (a component added, removed or swapped, compared with a baseline): helped 71%, no clear effect 12%, hurt 11%, mixed 6%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Vary the base Opt-OMD step size and combine learners with the LASU ensemble | CIFAR-10 and EuroSAT · average error (%) | Mixed | Label Shift Meets Online Learning: Ensuring Consistent Adaptation with Universal Dynamic Regret (CVPR 2025 · 0 citations) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 17 papers; from OpenAlex and the papers' affiliations).
- Nanjing University · 3 papers
- Tel Aviv University · 2 papers
- Apple · 2 papers
- Carnegie Mellon University · 2 papers
Use it
Download BibTeX of 29 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Constrained and unconstrained OCO algorithms (85 papers)
- Dynamic and adaptive regret (57 papers)
Elsewhere in the topic
- Online learning with feedback graphs (155 papers)
- Best-of-both-worlds bandit algorithms (125 papers)
- Learning for pricing and auctions (101 papers)
- Online decision-making with feedback structure (81 papers)
- Online allocation with constraints (77 papers)
- Learning-augmented online algorithms (73 papers)
- Last-iterate convergence in games (65 papers)
- Fast no-regret dynamics in games (37 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).