Research map / Learning theory and algorithms / Prediction and decision losses
PAC-Bayes Generalization Certificates for Learned Inductive Conformal Prediction
Apoorva Sharma, Sushant Veer, Asher Hancock, Heng Yang, Marco Pavone, Anirudha Majumdar
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 PAC-Bayes-constrained posterior optimization instead of split-based learned ICP | Learned conformal prediction · 1-D regression, N = 500 | prediction-set efficiency | Improved |
| Use PAC-Bayes-constrained posterior optimization instead of split-based learned ICP | Learned conformal prediction · Corrupted MNIST, N = 1000 | prediction set size | Improved |
| Fine-tune the score function on corrupted calibration data | Standard ICP · Corrupted MNIST | prediction set size | Improved |
| Use PAC-Bayes optimization rather than learned ICP calibrated with Vovk 2a | Learned ICP · Corrupted MNIST | prediction set size | Mixed |
| Use PAC-Bayes optimization rather than learned ICP calibrated with Vovk 2b | Learned ICP · Corrupted MNIST | prediction set size | Hurt |
| Use PAC-Bayes optimization rather than split-based learned ICP | Learned conformal prediction · Corrupted MNIST, varying data split ratio | prediction set size | Improved |
Closest papers by meaning
- Conformal Prediction with Learned Features (ICML 2024 · similarity 0.87)
- Verifiably Robust Conformal Prediction (NeurIPS 2024 · similarity 0.86)
- Direct Prediction Set Minimization via Bilevel Conformal Classifier Training (ICML 2025 · similarity 0.86)
- Optimal transport-based conformal prediction (ICML 2025 · similarity 0.85)
- Learning via Surrogate PAC-Bayes (NeurIPS 2024 · similarity 0.85)
- Backward Conformal Prediction (NeurIPS 2025 · similarity 0.85)
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.
