Research map / Label-efficient and robust learning / Recognition and long-tail
Participatory Personalization in Classification
Hailey Joren, Chirag Nagpal, Katherine A Heller, Berk Ustun
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 |
|---|---|---|---|
| let users opt out of personalization (Minimal vs 1Hot) (ablation) | personalized logistic regression · apnea | test error: 29.1 -> 28.9 | Improved |
| opt-out only (Minimal) on cardio_eicu (ablation) | 1Hot · cardio_eicu | test error: 21.4 -> 21.6 | No clear effect |
| assign multiple models over reporting options (Flat/Seq vs Minimal) (ablation) | Minimal system · apnea | test error: 28.9 -> 24.1 | Improved |
| assign multiple models over reporting options (Flat vs Minimal) (ablation) | Minimal system · cardio_eicu | test error: 21.6 -> 10.2 | Improved |
| sequential instead of flat reporting interface (ablation) | Flat system · apnea | test error: 24.1 -> 24.3 | No clear effect |
| sequential instead of flat reporting interface (ablation) | Flat system · coloncancer | test error: 36.6 -> 36.1 | Improved |
| replace opt-out with KNN imputation of group attributes (ablation) | IntHot static model · saps | test error: 20.7 -> 29.4 | Hurt |
| KNN imputation for opting out (ablation) | 1Hot/IntHot · 6 clinical datasets | rationality violations: 5/6 | Hurt |
| fit random forest instead of logistic regression | personalized model · saps | overall test error: 20.4 -> 14.1 | Mixed |
| switch metric to AUC and use Flat system | 1Hot logistic regression · cardio_eicu | test AUC: 0.864 -> 0.966 | Improved |
| random forest components in Flat system | 1Hot random forest · apnea | test error: 26.3 -> 12.2 | Improved |
| optimize for error vs AUC when pruning Seq | Seq on apnea · apnea | options pruned: 6 -> 4 | Mixed |
Closest papers by meaning
- Providing Fair Recourse over Plausible Groups (AAAI 2024 · similarity 0.83)
- Towards Safe Policy Learning under Partial Identifiability: A Causal Approach (AAAI 2024 · similarity 0.82)
- It's My Data Too: Private ML for Datasets with Multi-User Training Examples (ICML 2025 · similarity 0.82)
- I Prefer Not to Say: Protecting User Consent in Models with Optional Personal Data (AAAI 2024 · similarity 0.82)
- HYDRA: Model Factorization Framework for Black-Box LLM Personalization (NeurIPS 2024 · similarity 0.82)
- Learning from Aggregate responses: Instance Level versus Bag Level Loss Functions (ICLR 2024 · similarity 0.81)
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.
