Research map / Probabilistic and scientific modeling / Bayesian optimization
Distributionally Robust Bayesian Optimization with $\varphi$-divergences
Hisham Husain, Vu Nguyen, Anton van den Hengel
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 large epsilon_t in {0.5,1} when robust and stochastic optima differ (ablation) | DRBO · Benchmark functions, robust optimum differs from stochastic optimum | cumulative robust regret | Improved |
| Set epsilon_t toward 0 when robust and stochastic optima coincide (ablation) | DRBO · Benchmark functions, identical optima | optimization performance | Improved |
| Use adaptive epsilon_t instead of a fixed epsilon (ablation) | DRBO · Benchmark functions | optimization performance stability | Improved |
| Replace MMD adversary (linear programming) with closed-form phi-divergence reduction | DRBO · 5d benchmark function | computational cost per iteration: 5-times faster | Improved |
| Replace MMD adversary with closed-form phi-divergence reduction | DRBO · 6d benchmark function | computational cost per iteration: 10-times faster | Improved |
| Use chi2/TV DRBO instead of DRBO MMD | DRBO · Benchmark functions | cumulative robust regret | Improved |
| Use distributional robustness instead of vanilla BO ignoring context | BO · Benchmark functions | cumulative robust regret | Improved |
| Use distributional robustness instead of Stable-Opt worst-case over predefined contexts | Stable-Opt · Benchmark functions | cumulative robust regret | Improved |
| Increase context set size |C| from 30 to 100 (ablation) | DRBO chi2/TV/KL · Wind Power | optimization performance: 30 -> 100 | Improved |
| Increase context set size |C| from 100 to 500 (ablation) | DRBO chi2/TV/KL · Wind Power | optimization performance: 100 -> 500 | No clear effect |
| Increase context set size |C| (ablation) | DRBO MMD · Wind Power | optimization performance and cost | Mixed |
Closest papers by meaning
- Large-Scale Non-convex Stochastic Constrained Distributionally Robust Optimization (AAAI 2024 · similarity 0.90)
- Bayesian Nonparametrics Meets Data-Driven Distributionally Robust Optimization (NeurIPS 2024 · similarity 0.88)
- Distributionally Robust Optimization with Bias and Variance Reduction (ICLR 2024 · similarity 0.87)
- Wasserstein Distributionally Robust Optimization through the Lens of Structural Causal Models and Individual Fairness (NeurIPS 2024 · similarity 0.87)
- Outlier-Robust Distributionally Robust Optimization via Unbalanced Optimal Transport (NeurIPS 2024 · similarity 0.87)
- Differentiable Distributionally Robust Optimization Layers (ICML 2024 · similarity 0.87)
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.
