Research map / Optimization / Optimizers
A Model-Based Method for Minimizing CVaR and Beyond
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 |
|---|---|---|---|
| Replace stochastic subgradient with stochastic prox-linear (tighter truncated model) for CVaR | CVaR minimization · synthetic and real CVaR tasks, β=0.95 | robustness to step size | Helped |
| Use two separate regularization parameters with units-based defaults (SPL+) instead of a single step size (ablation) | SPL · synthetic smooth losses, β=0.95 | range of good step sizes λ | Helped |
| Use SPL+ with separate regularization instead of SGM | CVaR on absolute loss · synthetic absolute loss | sensitivity to λ | No clear effect |
| Use SPL+ with separate regularization instead of SGM | least squares on YearPredictionMSD and E2006-tfidf · real regression datasets | best achieved suboptimality | Mixed |
| Use SPL/SPL+ instead of SGM | logistic regression on Covertype and mushrooms · real classification datasets | suboptimality vs λ | Helped |
| Set λ=1 default for SPL+ via unit-balanced scaling | SPL+ · real regression datasets | best λ location | Helped |
Closest papers by meaning
- Boosting CVaR Policy Optimization with Quantile Gradients (ICML 2026 · similarity 0.89)
- Adaptive Sampling for Stochastic Risk-Averse Learning (NeurIPS 2020 · similarity 0.88)
- Risk-Sensitive Policy Optimization via Predictive CVaR Policy Gradient (ICML 2024 · similarity 0.87)
- Provably Efficient CVaR RL in Low-rank MDPs (ICLR 2024 · similarity 0.86)
- Return Capping: Sample Efficient CVaR Policy Gradient Optimisation (ICML 2025 · similarity 0.86)
- SOREL: A Stochastic Algorithm for Spectral Risks Minimization (ICLR 2025 · 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.