Uncertainty Quantification for Prior-Data Fitted Networks using Martingale Posteriors
Prior-data fitted networks (PFNs) have emerged as promising foundation models for prediction from tabular data sets, achieving state-of-the-art performance on small to moderate data sizes without tuning. While PFNs are motivated by Bayesian ideas, they do not provide any uncertainty quantification for predictive means, quantiles, or similar quantities. We propose a principled and efficient sampling procedure to construct Bayesian posteriors for such estimates based on Martingale posteriors, and prove its convergence. Several simulated and real-world data examples showcase the uncertainty quantification of our method in inference applications.
Code (0)
등록된 구현이 없습니다.
Tasks
Uncertainty QuantificationSimilar Papers 제목 키워드 기반
Frequentist Consistency of Prior-Data Fitted Networks for Causal Inference
Foundation models based on prior-data fitted networks (PFNs) have shown strong empirical performance in causal inference by framing the task as an in-context learning problem. However, it is unclear whether PFN-based cau…
Causal InferenceUncertainty quantification using martingales for misspecified Gaussian processes
We address uncertainty quantification for Gaussian processes (GPs) under misspecified priors, with an eye towards Bayesian Optimization (BO). GPs are widely used in BO because they easily enable exploration based on post…
Bayesian OptimizationGaussian ProcessesUncertainty QuantificationvalidTabMGP: Martingale Posterior with TabPFN
Bayesian inference provides principled uncertainty quantification but is often limited by the challenges of prior and likelihood elicitation. The martingale posterior (MGP) (Fong et al., 2023) offers an alternative by re…
Bayesian InferenceAsymptotically Log-Optimal Bayes-Assisted Confidence Sequences for Bounded Means
Confidence sequences based on test martingales provide time-uniform uncertainty quantification for the mean of bounded IID observations without parametric distributional assumptions. Their practical efficiency, however, …
Uncertainty Quantification for Inferring Hawkes Networks
Multivariate Hawkes processes are commonly used to model streaming networked event data in a wide variety of applications. However, it remains a challenge to extract reliable inference from complex datasets with uncertai…
Uncertainty Quantification