Stein variational gradient descent: A general purpose bayesian inference algorithm

Q Liu, D Wang - Advances in neural information processing …, 2016 - proceedings.neurips.cc
We propose a general purpose variational inference algorithm that forms a natural
counterpart of gradient descent for optimization. Our method iteratively transports a set of …

Scaling Hamiltonian Monte Carlo inference for Bayesian neural networks with symmetric splitting

AD Cobb, B Jalaian - Uncertainty in Artificial Intelligence, 2021 - proceedings.mlr.press
Abstract Hamiltonian Monte Carlo (HMC) is a Markov chain Monte Carlo (MCMC) approach
that exhibits favourable exploration properties in high-dimensional models such as neural …

It takes (only) two: Adversarial generator-encoder networks

D Ulyanov, A Vedaldi, V Lempitsky - … of the AAAI Conference on Artificial …, 2018 - ojs.aaai.org
We present a new autoencoder-type architecture that is trainable in an unsupervised mode,
sustains both generation and inference, and has the quality of conditional and unconditional …

Neutra-lizing bad geometry in hamiltonian monte carlo using neural transport

M Hoffman, P Sountsov, JV Dillon, I Langmore… - arXiv preprint arXiv …, 2019 - arxiv.org
Hamiltonian Monte Carlo is a powerful algorithm for sampling from difficult-to-normalize
posterior distributions. However, when the geometry of the posterior is unfavorable, it may …

Inference via low-dimensional couplings

A Spantini, D Bigoni, Y Marzouk - Journal of Machine Learning Research, 2018 - jmlr.org
We investigate the low-dimensional structure of deterministic transformations between
random variables, ie, transport maps between probability measures. In the context of …

Coupling techniques for nonlinear ensemble filtering

A Spantini, R Baptista, Y Marzouk - SIAM Review, 2022 - SIAM
We consider filtering in high-dimensional non-Gaussian state-space models with intractable
transition kernels, nonlinear and possibly chaotic dynamics, and sparse observations in …

Structured neural networks for density estimation and causal inference

A Chen, RI Shi, X Gao, R Baptista… - Advances in Neural …, 2024 - proceedings.neurips.cc
Injecting structure into neural networks enables learning functions that satisfy invariances
with respect to subsets of inputs. For instance, when learning generative models using …

Probability flow solution of the fokker–planck equation

NM Boffi, E Vanden-Eijnden - Machine Learning: Science and …, 2023 - iopscience.iop.org
The method of choice for integrating the time-dependent Fokker–Planck equation (FPE) in
high-dimension is to generate samples from the solution via integration of the associated …

Seismic tomography using variational inference methods

X Zhang, A Curtis - Journal of Geophysical Research: Solid …, 2020 - Wiley Online Library
Seismic tomography is a methodology to image the interior of solid or fluid media and is
often used to map properties in the subsurface of the Earth. In order to better interpret the …

[HTML][HTML] Sequential Monte Carlo with kernel embedded mappings: The mapping particle filter

M Pulido, PJ van Leeuwen - Journal of Computational Physics, 2019 - Elsevier
In this work, a novel sequential Monte Carlo filter is introduced which aims at an efficient
sampling of the state space. Particles are pushed forward from the prediction to the posterior …