Research map / Optimization / Nonconvex and smooth optimization
Stochastic Variance Reduced Primal Dual Algorithms for Empirical Composition Optimization
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 SVRPDA-I instead of the compared composition-optimization baselines | Risk-averse portfolio management · US Research Returns (18 datasets) | oracle calls to objective gap | Helped |
| Use SVRPDA-II instead of the compared composition-optimization baselines | Risk-averse portfolio management · US Research Returns (18 datasets) | oracle calls to objective gap | Helped |
| Approximate the reference inner-average gradient instead of storing it exactly (ablation) | SVRPDA-I, producing SVRPDA-II · Risk-averse portfolio management | oracle calls to objective gap | Hurt |
| Approximate the reference inner-average gradient instead of storing it exactly (ablation) | SVRPDA-I, producing SVRPDA-II · General empirical composition optimization | storage complexity: O(n_X d ℓ) -> O(d + n_X ℓ) | Helped |
| Use SVRPDA-I instead of C-SAGA | MDP policy evaluation · MDP policy evaluation (S = 10) | oracle calls to objective gap | Helped |
| Use SVRPDA-II instead of C-SAGA | MDP policy evaluation · MDP policy evaluation (S = 10 and S = 10^4) | oracle calls to objective gap | Mixed |
Closest papers by meaning
- Optimal Algorithms for Stochastic Multi-Level Compositional Optimization (ICML 2022 · similarity 0.90)
- Doubly Accelerated Stochastic Variance Reduced Dual Averaging Method for Regularized Empirical Risk Minimization (NeurIPS 2017 · similarity 0.89)
- Stochastic Primal-Dual Method for Empirical Risk Minimization with O(1) Per-Iteration Complexity (NeurIPS 2018 · similarity 0.89)
- A Near-Optimal Single-Loop Stochastic Algorithm for Convex Finite-Sum Coupled Compositional Optimization (ICML 2025 · similarity 0.87)
- Faster Stochastic Variance Reduction Methods for Compositional MiniMax Optimization (AAAI 2024 · similarity 0.87)
- Stability and Generalization of Stochastic Compositional Gradient Descent Algorithms (ICML 2024 · similarity 0.86)
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.