Newest papers

Cross-Fitting for Neural Posterior Estimation

Jeffrey Regier · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26733 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Neural Posterior Estimation (NPE) cocok dengan estimator kepadatan kondisional amortisasi ke set pelatihan simulasi dari realisasi data lengkap. Tujuan pelatihan NPE, log-probability, menguntungkan pendekatan belakang yang mengalokasikan kepadatan tinggi secara arbitrase ke titik pelatihan, dan keluarga khas estimator kepadatan kondisional amortisasi (misalnya, normalisasi aliran, Gaussian encoders) cukup fleksibel untuk memaksakan. Kebijakan overconfidence ekstrem yang dihasilkan diterjemahkan dengan buruk ke data yang ditahan. Stop awal dan regulator lainnya dapat mencegah degenerasi ini, tetapi mereka terbatas pada memilih pendekatan belakang di sepanjang trajektori optimalisasi yang tidak menguntungkan yang ditetapkan oleh residu sampel, di mana rata-rata belakang kurang cocok sementara variasi belakang berlebihan. Untuk menghindari per

Machine translation (Meta M2M100); the English original follows.

Abstract

Neural Posterior Estimation (NPE) fits an amortized conditional density estimator to a simulated training set of complete-data realizations. NPE's training objective, log-likelihood, favors posterior approximations that assign arbitrarily high density to the training points, and typical families of amortized conditional density estimators (e.g., normalizing flows, Gaussian encoders) are flexible enough to oblige. The resulting policy of extreme overconfidence translates poorly to held-out data. Early stopping and other regularizers can prevent this degeneracy, but they are limited to selecting posterior approximations along an unfavorable optimization trajectory set by in-sample residuals, where the posterior mean is underfit while the posterior variance is overfit. To avoid this trade-off, we propose Cross-Fitting NPE (CF-NPE), which uses sample splitting to separate conditional-mean estimation from the estimation of the posterior's higher moments. Using $K$-fold cross-fitting, CF-NPE fits $K$ posterior-mean regressors and then fits a conditional density estimator to the resulting out-of-fold residuals. In experiments with synthetic data and in a cosmological application, CF-NPE achieves higher held-out log-likelihood than standard NPE and typically improves calibration. CF-NPE is most useful in high-dimensional, simulation-limited settings.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Estimator, Mathematics, Density estimation, Algorithm, Gaussian, Set (abstract data type)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange