Research map / Bandits and online learning / Online convex optimization and games
Learning-augmented online algorithms
Use machine-learned predictions to improve online algorithms while preserving worst-case robustness.
73 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Online convex optimization and games, keyword cluster “price · pricing · auctions”
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.
21 papers in 2023–24 and 26 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic held steady, from 9.2% to 9.4%.
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 · Designing smoothing functions for improved worst-case competitive ratio in online optimization (NeurIPS 2016 · 17 citations)
- 2017 · Parameter-Free Online Learning via Model Selection (NeurIPS 2017 · 23 citations)
- 2018 · Competitive Caching with Machine Learned Advice (ICML 2018)
- 2020 · The Primal-Dual method for Learning Augmented Algorithms (NeurIPS 2020 · 32 citations)
- 2021 · Data-driven Competitive Algorithms for Online Knapsack and Set Cover (AAAI 2021 · 20 citations)
- 2022 · Robustification of Online Graph Exploration Methods (AAAI 2022 · 13 citations)
- 2023 · Learning-Augmented Algorithms for Online TSP on the Line (AAAI 2023 · 12 citations)
- 2024 · Learning-Augmented Priority Queues (NeurIPS 2024 · 2 citations)
- 2025 · Near-Optimal Consistency-Robustness Trade-Offs for Learning-Augmented Online Knapsack Problems (ICML 2025 · 1 citation)
- 2026 · Robust and Consistent Ski Rental with Distributional Advice (ICML 2026 · 0 citations)
Earliest papers here
- Designing smoothing functions for improved worst-case competitive ratio in online optimization (NeurIPS 2016 · 17 citations)
- Parameter-Free Online Learning via Model Selection (NeurIPS 2017 · 23 citations)
- Multi-Objective Non-parametric Sequential Prediction (NeurIPS 2017 · 2 citations)
What has been tried
45 papers here report 311 measured changes (a component added, removed or swapped, compared with a baseline): helped 64%, no clear effect 9%, hurt 19%, mixed 8%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| use separate past and predicted future costs rather than a summed condition | BP$(C, \beta, T)$, adversarial predictions · robustness | Helped | Learning-Augmented Algorithms for the Bahncard Problem (NeurIPS 2024 · 2 citations) | |
| Use θ=0, fully following predictions | Synthetic, Azure, and AppUsage traces with accurate predictions · average cost ratio | Helped | Learning-Augmented Online Algorithm for Two-Level Ski-Rental Problem (AAAI 2024 · 1 citation) | |
| Use f(i)=i^2 | Caching · smoothness | O(\log^2 k) \OPT; O(\sqrt{2\frac\eta\OPT}) | Mixed | Algorithms for Caching and MTS with reduced number of predictions (ICLR 2024 · 1 citation) |
| Randomize between RTC and Switch with the stated probability | Single-machine scheduling, perfect predictions · competitive ratio | 2 - \frac{B}{n} - \frac{2(1 - \frac{B}{n})(2 - \frac{B}{n})}{n+3 - \frac{2B}{n}} | Helped | Non-clairvoyant Scheduling with Partial Predictions (ICML 2024 · 1 citation) |
| Set ρ=1 instead of ρ=0 | synthetic worst-case instances · worst-case payoff ratio | Helped | Pareto-Optimality, Smoothness, and Stochasticity in Learning-Augmented One-Max-Search (ICML 2025 · 0 citations) | |
| disable sigma remapping | Synthetic hard known-IID instances, random order · competitive ratio | Hurt | Online bipartite matching with imperfect advice (ICML 2024 · 0 citations) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 47 papers; from OpenAlex and the papers' affiliations).
- CNRS · 3 papers
- University of Massachusetts Amherst · 3 papers
- National University of Singapore · 2 papers
- Bocconi University · 2 papers
- Google · 2 papers
- RelationalAI · 2 papers
- NTT · 2 papers
- The University of Tokyo · 2 papers
- University of Massachusetts at Amherst · 2 papers
- University of Oxford · 2 papers
Use it
Download BibTeX of 40 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
- Online learning with feedback graphs (155 papers)
- Best-of-both-worlds bandit algorithms (125 papers)
- Learning for pricing and auctions (101 papers)
- Online allocation with constraints (77 papers)
- Online decision-making with feedback structure (81 papers)
- Online linear-quadratic control regret (36 papers)
Elsewhere in the topic
- Constrained and unconstrained OCO algorithms (85 papers)
- Last-iterate convergence in games (65 papers)
- Dynamic and adaptive regret (57 papers)
- Parameter-free non-stationary online learning (45 papers)
- Fast no-regret dynamics in games (37 papers)
- Counterfactual regret minimization (15 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).