Conservative neural posterior estimation via distributionally robust training
Read the original on arXiv Machine Learning →The paper introduces DRO-NPE, a distributionally robust neural posterior estimation method that replaces the standard NPE objective with a worst‑case loss over a Wasserstein ambiguity set. It presents KL‑based metrics for miscoverage and miscalibration, demonstrating that the DRO‑NPE objective mitigates overfitting and reduces posterior overconfidence. Experiments on benchmark simulation‑based inference tasks show that DRO‑NPE consistently improves coverage and calibration while narrowing the gap between empirical and population NPE loss, especially in low‑simulation regimes.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.