Research map / Learning theory and algorithms / Prediction and decision losses
Multiple Robust Learning for Recommendation
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 |
|---|---|---|---|
| Use MR instead of the strongest reported baseline | MF recommendation backbone · Coat | AUC: 0.7157 -> 0.7356 | Helped |
| Use MR instead of the strongest reported baseline | MF recommendation backbone · Yahoo | nDCG@10: 0.7738 -> 0.7833 | Helped |
| Use MR instead of the strongest reported baseline | NCF recommendation backbone · Yahoo | nDCG@10: 0.7864 -> 0.8112 | Helped |
| Use MR rather than the strongest baseline | recommendation model · ML-100k-1 | nDCG@10: 0.3536 -> 0.3622 | Helped |
| Use MR rather than the strongest baseline | recommendation model · ML-100k-2 | AUC: 0.6440 -> 0.6968 | Helped |
| Use MR rather than the strongest baseline | recommendation model · ML-100k-3 | nDCG@10: 0.2838 -> 0.3100 | Helped |
| Increase the number of MF imputation models from one to four (ablation) | MR on Coat with MF prediction backbone · Coat | AUC: 0.7209 -> 0.7435 | Helped |
| Increase the number of MF imputation models from four to five (ablation) | MR on Coat with MF prediction backbone · Coat | AUC: 0.7435 -> 0.7421 | Hurt |
| Use two NCF imputations instead of one (ablation) | MR on Coat · Coat | AUC: 0.7230 -> 0.7381 | Helped |
| Use two MF and NCF imputations instead of one MF and one NCF imputation (ablation) | MR on Coat · Coat | AUC: +3.4‰ | Helped |
| Use different propensity model types together (ablation) | MR on Coat with MF imputation and prediction models · Coat | AUC: 0.7219 -> 0.7370 | Helped |
| Add another same-type NB propensity model (ablation) | MR on Coat with MF imputation and prediction models · Coat | AUC: 0.7219 -> 0.7216 | Hurt |
Closest papers by meaning
- Relaxing the Accurate Imputation Assumption in Doubly Robust Learning for Debiased Collaborative Filtering (ICML 2024 · similarity 0.91)
- LLM4RSR: Large Language Models as Data Correctors for Robust Sequential Recommendation (AAAI 2025 · similarity 0.86)
- Fine-Grained Dynamic Framework for Bias-Variance Joint Optimization on Data Missing Not at Random (NeurIPS 2024 · similarity 0.85)
- Addressing Hidden Confounding with Heterogeneous Observational Datasets for Recommendation (NeurIPS 2024 · similarity 0.84)
- Unbiased Recommender Learning from Implicit Feedback via Weakly Supervised Learning (ICML 2025 · similarity 0.83)
- Advancing Loss Functions in Recommender Systems: A Comparative Study with a Rényi Divergence-Based Solution (AAAI 2025 · 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.