Research map / Optimization / Nonconvex and smooth optimization
Variance-reduced policy gradient
Applies variance reduction to policy gradient and policy evaluation in reinforcement learning.
17 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2017–2026) · an approach in Nonconvex and smooth optimization, keyword cluster “variance · reduction · reduced”
State History Tried Preprints Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
2 papers in 2023–24 and 4 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 0.6% to 1.1%.
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
- 2017 · Stochastic Variance Reduction Methods for Policy Evaluation (ICML 2017 · 62 citations)
- 2018 · Stochastic Variance-Reduced Policy Gradient (ICML 2018 · 13 citations)
- 2019 · SAGA with Arbitrary Sampling (ICML 2019 · 14 citations)
- 2020 · Reanalysis of Variance Reduced Temporal Difference Learning (ICLR 2020 · 16 citations)
- 2022 · Bregman Gradient Policy Optimization (ICLR 2022 · 1 citation)
- 2023 · A Smooth Binary Mechanism for Efficient Private Continual Observation (NeurIPS 2023 · 2 citations)
- 2025 · Nearly Optimal Sample Complexity for Learning with Label Proportions (ICML 2025 · 0 citations)
- 2026 · Variance-Reduced $(\varepsilon, \delta)-$Unlearning using Forget Set Gradients (ICML 2026)
Earliest papers here
- Stochastic Variance Reduction Methods for Policy Evaluation (ICML 2017 · 62 citations)
- Stochastic Variance-Reduced Policy Gradient (ICML 2018 · 13 citations)
- SAGA with Arbitrary Sampling (ICML 2019 · 14 citations)
What has been tried
15 papers here report 113 measured changes (a component added, removed or swapped, compared with a baseline): helped 77%, no clear effect 12%, hurt 5%, mixed 5%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| remove clipping from the proposed loss | learning experiments · test accuracy | No clear effect | Nearly Optimal Sample Complexity for Learning with Label Proportions (ICML 2025 · 0 citations) |
Recent preprints
Loading recent arXiv preprints…
Use it
Download BibTeX of 17 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
- Stochastic variance reduction (86 papers)
- Proximal finite-sum methods (47 papers)
Elsewhere in the topic
- Adaptive first-order convex methods (190 papers)
- Coordinate and Adam convergence analysis (188 papers)
- Stochastic gradient last-iterate analysis (178 papers)
- Nonconvex minimax and constrained methods (100 papers)
- Stochastic convex optimization complexity (94 papers)
- Acceleration and Nesterov dynamics (91 papers)
- Equilibrium learning in zero-sum games (88 papers)
- Complexity of nonconvex first-order methods (79 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).