Research map / Learning theory and algorithms / Prediction and decision losses
Learning Prescriptive ReLU Networks
Wei Sun, Asterios Tsiourvas
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 P-ReLU instead of the strongest reported benchmark | prescription learning on synthetic datasets · Synthetic Datasets 1, 2, 5, and 6 | prescription accuracy (%): 76.03 ± 3.55 -> 84.47 ± 2.50; 60.59 ± 0.70 -> 66.72 ± 2.32; 82.19 ± 1.35 -> 88.80 ± 1.42; 81.13 ± 1.29 -> 87.08 ± 2.14 | Improved |
| Use P-ReLU instead of the strongest reported benchmark | prescription learning on Synthetic Dataset 3 · Synthetic Dataset 3 | prescription accuracy (%): 100 ± 0.00 -> 100 ± 0.00 | No clear effect |
| Use P-ReLU instead of the strongest reported benchmark | prescription learning on Synthetic Dataset 4 · Synthetic Dataset 4 | prescription accuracy (%): 93.71 ± 0.31 -> 93.75 ± 1.18 | No clear effect |
| Use P-ReLU instead of R&C LR | Warfarin prescription learning · Warfarin prescription | prescription accuracy (%): 66.68±0.67 -> 68.27 ± 0.76 | Improved |
| Add sparse one-layer architecture and magnitude-based weight pruning (ablation) | P-ReLU for Warfarin prescription · Warfarin prescription, unconstrained | prescription accuracy (%) | Hurt |
| Add the BMI-based treatment constraint using M=1,000 (ablation) | one-layer P-ReLU for Warfarin prescription · Warfarin prescription, constrained | prescription accuracy (%): 64.48 % -> 64.33% | Hurt |
| Train P-ReLU instead of OPT | Warfarin prescription · Warfarin prescription | training runtime: ≈2.6 mins -> ≈8 secs | Improved |
Closest papers by meaning
- Activation-Descent Regularization for Input Optimization of ReLU Networks (ICML 2024 · similarity 0.85)
- Expressivity of ReLU-Networks under Convex Relaxations (ICLR 2024 · similarity 0.85)
- Simplicity Bias and Optimization Threshold in Two-Layer ReLU Networks (ICML 2025 · similarity 0.84)
- Make Haste Slowly: A Theory of Emergent Structured Mixed Selectivity in Feature Learning ReLU Networks (ICLR 2025 · similarity 0.84)
- RePO: Understanding Preference Learning Through ReLU-Based Optimization (NeurIPS 2025 · similarity 0.84)
- Towards Safe Policy Learning under Partial Identifiability: A Causal Approach (AAAI 2024 · similarity 0.83)
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.
