Research map / Learning theory and algorithms / Submodular and game algorithms
Robust Optimization for Non-Convex Objectives
Robert S. Chen, Brendan Lucier, Yaron Singer, Vasilis Syrgkanis
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 Composite Method instead of the strongest baseline | Robust neural network training · Background Set | Individual Bottleneck Loss: 2.06 -> 1.31 | Improved |
| Use Composite Method instead of the strongest baseline | Robust neural network training · Shrink Set | Individual Bottleneck Loss: 1.74 -> 1.30 | Improved |
| Use Composite Method instead of the strongest baseline | Robust neural network training · Pixel Set | Individual Bottleneck Loss: 1.30 -> 1.25 | Improved |
| Use Composite Method instead of the strongest baseline | Robust neural network training · Mixed Set | Individual Bottleneck Loss: 1.46 -> 1.25 | Improved |
| Replace Hybrid Method with Composite Method (ablation) | Robust neural network training · Background Set | Individual Bottleneck Loss: 1.38 -> 1.31 | Improved |
| Replace Hybrid Method with Composite Method (ablation) | Robust neural network training · Shrink Set | Individual Bottleneck Loss: 1.48 -> 1.30 | Improved |
| Replace Hybrid Method with Composite Method (ablation) | Robust neural network training · Pixel Set | Individual Bottleneck Loss: 1.29 -> 1.25 | Improved |
| Replace Hybrid Method with Composite Method (ablation) | Robust neural network training · Mixed Set | Individual Bottleneck Loss: 1.36 -> 1.25 | Improved |
| Use robust optimization instead of the strongest baseline | Influence maximization · Wikipedia A | Bottleneck Influence: 83.35 -> 94.33 | Improved |
| Use robust optimization instead of the strongest baseline | Influence maximization · Wikipedia B | Bottleneck Influence: 48.92 -> 66.42 | Improved |
| Use robust optimization instead of the strongest baseline | Influence maximization · Complete A | Bottleneck Influence: 21.99 -> 36.34 | Improved |
| Use robust optimization instead of the strongest baseline | Influence maximization · Complete B | Bottleneck Influence: 11.27 -> 17.91 | Improved |
Closest papers by meaning
- Distributionally Robust Optimization with Bias and Variance Reduction (ICLR 2024 · similarity 0.90)
- Stochastic Gradient Methods for Distributionally Robust Optimization with f-divergences (NeurIPS 2016 · similarity 0.89)
- Outlier-Robust Distributionally Robust Optimization via Unbalanced Optimal Transport (NeurIPS 2024 · similarity 0.87)
- Distributionally Robust Performative Optimization (NeurIPS 2025 · similarity 0.86)
- Drago: Primal-Dual Coupled Variance Reduction for Faster Distributionally Robust Optimization (NeurIPS 2024 · similarity 0.86)
- Bayesian Nonparametrics Meets Data-Driven Distributionally Robust Optimization (NeurIPS 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.
