Search Results for author:
Found 10 papers, 7 papers with code
Persistent Sampling: Unleashing the Potential of Sequential Monte Carlo
We introduce persistent sampling (PS), an extension of SMC that mitigates this issue by allowing particles from previous iterations to persist.
Sequential Kalman Tuning of the $t$-preconditioned Crank-Nicolson algorithm: efficient, adaptive and gradient-free inference for Bayesian inverse problems
In this work we propose embedding EKI and Flow Annealed Kalman Inversion (FAKI), its normalizing flow (NF) preconditioned variant, within a Bayesian annealing scheme as part of an adaptive implementation of the $t$-preconditioned Crank-Nicolson (tpCN) sampler.
Flow Annealed Kalman Inversion for Gradient-Free Inference in Bayesian Inverse Problems
For many scientific inverse problems we are required to evaluate an expensive forward model.
Fluctuation without dissipation: Microcanonical Langevin Monte Carlo
In contrast to the common belief, we show that the fluctuation-dissipation theorem is not required because only the configuration space distribution, and not the full phase space distribution, needs to be canonical.
Probabilistic Autoencoder
The PAE is fast and easy to train and achieves small reconstruction errors, high sample quality, and good performance in downstream tasks.
Ranked #1 on
Out-of-Distribution Detection
on Fashion-MNIST
Normalizing Constant Estimation with Gaussianized Bridge Sampling
Normalizing constant (also called partition function, Bayesian evidence, or marginal likelihood) is one of the central goals of Bayesian inference, yet most of the existing methods are both expensive and inaccurate.
Uncertainty Quantification with Generative Models
We develop a generative model-based approach to Bayesian inverse problems, such as image reconstruction from noisy and incomplete images.
Kepler data analysis: non-Gaussian noise and Fourier Gaussian process analysis of star variability
We develop a statistical analysis model of Kepler star flux data in the presence of planet transits, non-Gaussian noise, and star variability.
Earth and Planetary Astrophysics
Perturbation theory, effective field theory, and oscillations in the power spectrum
Shell crossings lead to non-perturbative effects, and the PT ignorance can be quantified in terms of their ratio, which is also the transfer function squared in the absence of stochasticity.
Cosmology and Nongalactic Astrophysics
Lagrangian perturbation theory at one loop order: successes, failures, and improvements
We find that the analytic calculations are in a good agreement with the LPT simulations, but when compared to full N-body simulations, we find that while one loop calculations improve upon the Zel'dovich approximation in the power spectrum, they still significantly lack power.
Cosmology and Nongalactic Astrophysics
