Jeffrey Regier · 2026 · arXiv
DOI: 10.48550/arXiv.2610.26733 · open-access PDF
Neural Posterior Estimation (NPE) se ajusta a un estimador de densidad condicional amortizado a un conjunto de entrenamiento simulado de realizaciones de datos completos. El objetivo de entrenamiento de NPE, log-probelihood, favorece aproximaciones posteriores que asignan densidad arbitrariamente alta a los puntos de entrenamiento, y las familias típicas de estimadores de densidad condicional amortizados (por ejemplo, flujos normalizantes, codificadores de Gauss) son lo suficientemente flexibles como para obligar. La política resultante de sobreconfianza extrema se traduce mal en datos detenidos.
Machine translation (Meta M2M100); the English original follows.
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
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.26733, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange