Research map / Probabilistic and scientific modeling / Treatment effects
Direct Heterogeneous Causal Learning for Resource Allocation Problems in Marketing
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 |
|---|---|---|---|
| directly learn treatment effect as the decision factor | DUM · CRITEO-UPLIFT v2 | AUUC: 0.8440 ± 0.0054 -> 0.8749 ± 0.0155 | Helped |
| directly learn treatment effect as the decision factor | DUM · Marketing data | AUUC: 0.7826 ± 0.0077 -> 0.8077 ± 0.0003 | Helped |
| use X-Learner instead of S-Learner | cost-unaware binary treatment assignment · CRITEO-UPLIFT v2 | AUUC: 0.8440 ± 0.0054 -> 0.8304 ± 0.0377 | Hurt |
| use Causal Forest instead of S-Learner | cost-unaware binary treatment assignment · Marketing data | AUUC: 0.7826 ± 0.0077 -> 0.7468 ± 0.0059 | Hurt |
| directly predict ROI rather than combine predicted revenue and cost | DRP · CRITEO-UPLIFT v2 | AUCC: 0.7561 ± 0.0113 -> 0.7739 ± 0.0002 | Helped |
| directly predict ROI rather than combine predicted revenue and cost | DRP · Marketing data | AUCC: 0.7153 ± 0.0142 -> 0.7666 ± 0.0003 | Helped |
| replace direct ROI prediction with Direct Rank | cost-aware binary treatment assignment · Marketing data | AUCC: 0.7666 ± 0.0003 -> 0.7414 ± 0.0101 | Hurt |
| directly predict marginal utilities and use the equivalent allocation algorithm | DPM · Marketing data | MT-AUCC: 0.7250 ± 0.0143 -> 0.7860 ± 0.0007 | Helped |
| directly predict marginal utilities instead of using causal-forest predictions in a two-phase method | DPM · Marketing data | MT-AUCC: 0.7395 ± 0.0159 -> 0.7860 ± 0.0007 | Helped |
| deploy DPM rather than TPM-CF | Meituan online A/B test · Meituan online A/B test | incremental orders: 14.3% | Helped |
| deploy DPM rather than TPM-CF | Meituan online A/B test · Meituan online A/B test | GMV: 13.6% | Helped |
Closest papers by meaning
- Bi-Level Decision-Focused Causal Learning for Large-Scale Marketing Optimization: Bridging Observational and Experimental Data (NeurIPS 2025 · similarity 0.90)
- FedRSPO+: A Heterogeneity-aware Algorithm for Decision-focused Federated Learning (NeurIPS 2026 · similarity 0.82)
- Pricing with Contextual Elasticity and Heteroscedastic Valuation (ICML 2024 · similarity 0.82)
- Transfer Learning for Nonparametric Contextual Dynamic Pricing (ICML 2025 · similarity 0.82)
- Invariant Deep Uplift Modeling for Incentive Assignment in Online Marketing via Probability of Necessity and Sufficiency (ICML 2025 · similarity 0.82)
- Learning to price with resource constraints: from full information to machine-learned prices (NeurIPS 2025 · similarity 0.82)
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.