A numerical polology sampler maps ghost- and tachyon-free regions of bosonic EFT coupling spaces and finds new particle branches, including a spin-one mode in general rank-2 tensor theories.
hub
BlackJAX: composable Bayesian inference in JAX
19 Pith papers cite this work, alongside 8 external citations. Polarity classification is still indexing.
19
Pith papers citing it
8
external citations · external index
hub tools
citation-role summary
background 1
method 1
citation-polarity summary
representative citing papers
Scalable Torus Graph models with stochastic inference allow large-scale analysis of neural phase couplings and their dynamics in LFP data.
gemlib.mcmc supplies composable kernel abstractions for Metropolis-within-Gibbs sampling via writer monads, allowing concise expression and reuse of complex MCMC algorithms for partially observed epidemic models.
Direct sampling under a uniform-in-dL prior raises P(H0>120) from 0.017 to 0.159 for GW170817; post-hoc reweighting recovers only 0.041 because a low-dL mode is undersampled.
SchwarMAX delivers a fast GPU-native Schwarzschild modeling code that recovers density profiles and bar pattern speed from mock IFU data of a simulated barred galaxy.
Joint 2nd- and 3rd-order cosmic shear analysis on KiDS-Legacy data produces Ω_m = 0.297^{+0.056}_{-0.040} and S_8 = 0.806^{+0.025}_{-0.023}, consistent with Planck and prior KiDS results while improving Ω_m precision.
Flow-ABI trains flow-matching models on historical data to produce a set-conditioned functional posterior sampler that delivers near-real-time Bayesian inference for regression and inverse PDE tasks without per-observation optimization.
GenSBI delivers JAX-native implementations of generative SBI methods with transformer backbones and reports near-ideal calibration scores on standard benchmarks.
Tempered sequential Monte Carlo samples from a Boltzmann-tilted distribution over controllers to optimize trajectories and policies under differentiable dynamics.
E-value sequential tests enable early stopping of MCMC sampling in Bayesian deep ensembles, often needing only a fraction of the full budget while improving over standard deep ensembles.
Time delay likelihoods modeled with Gaussian processes develop a boundary-driven W-shape with a global maximum at the true delay and rises at observation window edges, misleading nested sampling and biasing H0 high.
Dorito enables diffraction-limited image reconstruction from JWST AMI observations by deconvolving images or Fourier observables using maximum entropy and total variation regularization.
Empirical-Bayes hierarchical unfolding for gamma-ray spectra with Poisson ON/OFF likelihood, adaptive Richardson-Lucy prior, and NUTS posterior sampling, yielding spectra consistent with frequentist regularized ML.
PSD modeling of SFR scatter at six timescales shows dominant variability on 10-30 Myr scales, stronger in lower-mass galaxies at z=3-8.
An importance sampling correction is added to integrated Laplace approximation so that the approximate posterior for latent Gaussian models converges to the true posterior as the number of samples grows.
bde is a new Python package that implements Bayesian deep ensembles via efficient JAX-based Microcanonical Langevin Ensembles for tabular regression and classification with uncertainty estimates.
Numerical simulations benchmark the eikonal and post-Kerr approximations for quasinormal modes in deformed Kerr spacetimes, quantifying their errors relative to expected observational precision.
Bayesian evidence prefers a low-redshift supernova magnitude offset over dynamical dark energy when DES-5Y is combined with DESI BAO, but only under the assumption that Lambda CDM is correct.
Mass of 13.7 Earth masses and density 0.4 g cm^{-3} measured for TOI-1883 b, a super-Neptune in the ridge regime around an early-M dwarf, with implications for disk migration and photoevaporation.
citing papers explorer
-
Numerical polology: towards next-generation model-building for cosmology
A numerical polology sampler maps ghost- and tachyon-free regions of bosonic EFT coupling spaces and finds new particle branches, including a spin-one mode in general rank-2 tensor theories.
-
Torus Graphs for Large Scale Neural Phase Analysis
Scalable Torus Graph models with stochastic inference allow large-scale analysis of neural phase couplings and their dynamics in LFP data.
-
gemlib.mcmc: composable kernels for Metropolis-within-Gibbs sampling schemes
gemlib.mcmc supplies composable kernel abstractions for Metropolis-within-Gibbs sampling via writer monads, allowing concise expression and reuse of complex MCMC algorithms for partially observed epidemic models.
-
Rapid Hubble constant inference from GW170817 using GPU-accelerated nested sampling: prior sensitivity and the limits of post-hoc reweighting
Direct sampling under a uniform-in-dL prior raises P(H0>120) from 0.017 to 0.159 for GW170817; post-hoc reweighting recovers only 0.041 because a low-dL mode is undersampled.
-
SchwarMAX: a GPU-friendly Schwarzschild orbit-superposition modelling framework
SchwarMAX delivers a fast GPU-native Schwarzschild modeling code that recovers density profiles and bar pattern speed from mock IFU data of a simulated barred galaxy.
-
KiDS-Legacy: Joint analysis of second- and third-order cosmic shear
Joint 2nd- and 3rd-order cosmic shear analysis on KiDS-Legacy data produces Ω_m = 0.297^{+0.056}_{-0.040} and S_8 = 0.806^{+0.025}_{-0.023}, consistent with Planck and prior KiDS results while improving Ω_m precision.
-
Flow-based generative models for amortized Bayesian inference in regression and inverse PDE problems
Flow-ABI trains flow-matching models on historical data to produce a set-conditioned functional posterior sampler that delivers near-real-time Bayesian inference for regression and inverse PDE tasks without per-observation optimization.
-
GenSBI: Generative Methods for Simulation-Based Inference in JAX
GenSBI delivers JAX-native implementations of generative SBI methods with transformer backbones and reports near-ideal calibration scores on standard benchmarks.
-
Tempered Sequential Monte Carlo for Trajectory and Policy Optimization with Differentiable Dynamics
Tempered sequential Monte Carlo samples from a Boltzmann-tilted distribution over controllers to optimize trajectories and policies under differentiable dynamics.
-
Towards E-Value Based Stopping Rules for Bayesian Deep Ensembles
E-value sequential tests enable early stopping of MCMC sampling in Bayesian deep ensembles, often needing only a fraction of the full budget while improving over standard deep ensembles.
-
Global structure of the time delay likelihood
Time delay likelihoods modeled with Gaussian processes develop a boundary-driven W-shape with a global maximum at the true delay and rises at observation window edges, misleading nested sampling and biasing H0 high.
-
Image reconstruction with the JWST Interferometer
Dorito enables diffraction-limited image reconstruction from JWST AMI observations by deconvolving images or Fourier observables using maximum entropy and total variation regularization.
-
Empirical-Bayes Unfolding of $\gamma$-ray Spectra
Empirical-Bayes hierarchical unfolding for gamma-ray spectra with Poisson ON/OFF likelihood, adaptive Richardson-Lucy prior, and NUTS posterior sampling, yielding spectra consistent with frequentist regularized ML.
-
Star-formation variability on the star-forming main sequence during the Epoch of Reionization
PSD modeling of SFR scatter at six timescales shows dominant variability on 10-30 Myr scales, stronger in lower-mass galaxies at z=3-8.
-
Corrected Integrated Laplace Approximation for Bayesian Inference in Latent Gaussian Models
An importance sampling correction is added to integrated Laplace approximation so that the approximate posterior for latent Gaussian models converges to the true posterior as the number of samples grows.
-
bde: A Python Package for Bayesian Deep Ensembles via MILE
bde is a new Python package that implements Bayesian deep ensembles via efficient JAX-based Microcanonical Langevin Ensembles for tabular regression and classification with uncertainty estimates.
-
Confronting eikonal and post-Kerr methods with numerical evolution of scalar field perturbations in spacetimes beyond Kerr
Numerical simulations benchmark the eikonal and post-Kerr approximations for quasinormal modes in deformed Kerr spacetimes, quantifying their errors relative to expected observational precision.
-
Dynamic or Systematic? Bayesian model selection between dark energy and supernova biases
Bayesian evidence prefers a low-redshift supernova magnitude offset over dynamical dark energy when DES-5Y is combined with DESI BAO, but only under the assumption that Lambda CDM is correct.
-
The mass of TOI-1883 b: A low density super-Neptune in the ridge regime transiting an early-M dwarf
Mass of 13.7 Earth masses and density 0.4 g cm^{-3} measured for TOI-1883 b, a super-Neptune in the ridge regime around an early-M dwarf, with implications for disk migration and photoevaporation.