paper-with-me

Papers

Flow-Transformed Implicit Processes for Function-Space Variational Inference

2026-06-01 · Luis A. Ortega, Andrés R. Masegosa, Thomas D. Nielsen arxiv

Implicit-process priors define distributions over functions through flexible generative mechanisms, making them attractive for Bayesian function-space modelling. However, performing posterior inference with such priors is challenging because their induced function-space distributions are typically not available in closed form. One practical strategy is to approximate the prior using a finite collection of sampled functions, and then represent posterior functions as learned combinations of these samples. Existing approaches commonly place a Gaussian variational distribution over the combination weights. While tractable, this choice limits the shapes of posterior uncertainty that can be represented, especially when the true posterior is asymmetric, heavy-tailed, or multimodal. We propose Flow-Transformed Implicit Processes (FTIP), a variational inference method that makes this finite-dimensional function-space approximation more expressive. Instead of using a Gaussian distribution over the combination weights, FTIP uses a normalizing flow to define a richer variational distribution. This induces a flexible posterior distribution over functions while preserving tractable optimization. We train the model using a Black-Box α objective, allowing us to compare mass-covering and mode-seeking variational behaviour. Experiments show that FTIP captures asymmetric and multimodal posterior structure in function space that Gaussian coefficient approximations tend to smooth or collapse.

📄 PDF Abstract BibTeX arXiv:2606.01954

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

A Unifying View on Implicit Bias in Training Linear Neural Networks

2020-10-06 · ICLR 2021 1 · Chulhee Yun, Shankar Krishnan, Hossein Mobahi

We study the implicit bias of gradient flow (i.e., gradient descent with infinitesimal step size) on linear neural network training. We propose a tensor formulation of neural networks that includes fully-connected, diago…

Tensor Networks

Efficient Transformed Gaussian Process State-Space Models for Non-Stationary High-Dimensional Dynamical Systems

2025-03-24 · Zhidi Lin, Ying Li, Feng Yin, Juan Maroñas 외

Gaussian process state-space models (GPSSMs) offer a principled framework for learning and inference in nonlinear dynamical systems with uncertainty quantification. However, existing GPSSMs are limited by the use of mult…

Computational EfficiencyGaussian ProcessesState EstimationState Space Models+3

HFNO: an interpretable data-driven decomposition strategy for turbulent flows

2025-11-03 · Marco Cayuela, Vincent Le Chenadec, Peter Schmid, Taraneh Sayadi arxiv

Fourier Neural Operators (FNOs) have demonstrated exceptional accuracy in mapping functional spaces by leveraging Fourier transforms to establish a connection with underlying physical principles. However, their opaque in…

Approximate inference with Wasserstein gradient flows

2018-06-12 · Charlie Frogner, Tomaso Poggio

We present a novel approximate inference method for diffusion processes, based on the Wasserstein gradient flow formulation of the diffusion. In this formulation, the time-dependent density of the diffusion is derived as…

Simplicity bias in the parameter-function map of deep neural networks

2019-05-28 · Guillermo Valle-Pérez, Chico Q. Camargo, Ard A. Louis

The idea that neural networks may exhibit a bias towards simplicity has a long history. Simplicity bias provides a way to quantify this intuition. It predicts, for a broad class of input-output maps which can describe m…