New study on No-U-Turn Sampler for accelerated mixing in Hamiltonian Monte Carlo.
problem Achieving accelerated convergence in Hamiltonian Monte Carlo.
method Combining concentration of measure and coupling analysis for mixing.
result Rigorous mixing guarantees for the No-U-Turn Sampler in certain Gaussian distributions.
NUTS mixing time scales as d^(1/4) for Gaussian distributions.
problem Improving the efficiency of the No-U-Turn Sampler (NUTS) for Gaussian distributions.
method Coupling argument leveraging geometric structure of Gaussian concentration, uniformity analysis of NUTS transitions.
result The mixing time of NUTS scales as d^(1/4) for Gaussian distributions, up to logarithmic factors.
This paper analyzes the convergence of dynamic HMC and NUTS methods.
problem Theoretical understanding of dynamic HMC and NUTS convergence.
method General class of MCMC algorithms, NUTS as a particular case, geometric ergodicity, irreducibility.
result NUTS is geometrically ergodic under certain conditions and ergodic without bounded stepsize.
WALNUTS improves sampling efficiency and robustness for multi-scale distributions.
problem Adapting leapfrog step size for multi-scale posterior distributions.
method Adapts leapfrog step size at fixed intervals of simulated time, selecting the largest step size to keep energy error below a threshold.
result Substantial improvements in sampling efficiency and robustness compared to standard NUTS.
This paper compares two NUTS variants and analyzes their convergence and mixing times.
problem Theoretical comparison and convergence guarantees of NUTS variants.
method Deriving necessary and sufficient conditions for geometric ergodicity, and analyzing mixing times.
result NUTS-mul and NUTS-BPS have nearly identical qualitative behavior but differ quantitatively in convergence rates.
L-HNNs improve Bayesian inference efficiency by reducing gradient computation.
problem Efficient Bayesian inference with minimal gradient computation.
method Integrating L-HNNs into NUTS with online error monitoring.
result L-HNNs in NUTS outperform NUTS in complex posterior densities.
ATLAS adapts HMC step size and trajectory length for complex geometries.
problem Sampling complex geometries with constant step size HMC/NUTS.
method Adapts step size and trajectory length using local Hessian and no U-turn condition.
result ATLAS accurately samples complex geometries, outperforming NUTS.
GIST adapts HMC by tuning parameters based on position and momentum.
problem Locally adaptive sampling in Hamiltonian Monte Carlo.
method GIST uses Gibbs sampling to adaptively tune HMC parameters.
result GIST improves sampling efficiency for high-dimensional models.
PDMP samplers improve Bayesian PDE coefficient inference.
problem Efficient Bayesian inference in non-linear inverse problems with expensive likelihoods.
method Piecewise deterministic Markov process (PDMP) with surrogate-assisted thinning.
result PDMP samplers achieve higher accuracy and efficiency than traditional methods.
ChEES-HMC improves SMC samplers' efficiency and speed.
problem Efficiently sampling from complex posterior distributions.
method Incorporating ChEES-HMC into SMC samplers.
result ChEES-HMC outperforms NUTS in speed and efficiency.
We explore a general framework in Markov chain Monte Carlo (MCMC) sampling where sequential proposals are tried as a candidate for the next state of the Markov chain. This sequential-proposal framework can be applied to various existing MCMC methods, including Metropolis-Hastings algorithms using random proposals and m…
We describe a simple, low-level approach for embedding probabilistic programming in a deep learning ecosystem. In particular, we distill probabilistic programming down to a single abstraction---the random variable. Our lightweight implementation in TensorFlow enables numerous applications: a model-parallel variational …
NumPyro is a lightweight library that provides an alternate NumPy backend to the Pyro probabilistic programming language with the same modeling interface, language primitives and effect handling abstractions. Effect handlers allow Pyro's modeling API to be extended to NumPyro despite its being built atop a fundamentall…
Latent-IMH improves Bayesian inference for expensive operators.
problem Efficient sampling from posterior distributions in inverse problems with computationally expensive operators.
method Metropolis-Hastings independence sampler using approximate and exact operators.
result Latent-IMH outperforms existing methods in computational efficiency.
New variational flows improve Monte Carlo and normalization tasks.
problem Intractable global optimum in expressive variational families.
method Constructing asymptotically exact variational flows from involutive MCMC kernels.
result Provable total variation convergence of new variational families.
Improved particle filters for estimating model parameters using differentiable resampling.
problem Inability to differentiate sampling and resampling steps in particle filters.
method Extended reparameterisation trick to include stochastic input, enabling differentiation. Used p-MCMC and NUTS for parameter estimation.
result NUTS improves mixing of Markov chain and produces more accurate results in less time.
U-turn chains improve sampling from complex distributions.
problem Sampling from high-dimensional learned distributions.
method Iterative forward-backward diffusion steps with Metropolis-Hastings correction.
result Minimal U-turn dynamics exhibit phase transitions and layer-ordering inversion.
Novel framework for efficient Gaussian process models with monotonicity constraints.
problem Improving predictive accuracy and reducing uncertainty in high-dimensional problems with monotonicity constraints.
method Virtual point-based framework using regularized linear randomize-then-optimize (RLRTO) and No U-Turn Sampler (NUTS) for efficient sampling.
result Significant improvements in computational efficiency with the RLRTO method and NUTS enhancements.
New Langevin method achieves third order convergence for strongly log-concave distributions.
problem Sampling from complex distributions efficiently.
method Underdamped Langevin diffusion with third order convergence.
result Achieves 2-Wasserstein error of ε in O(√d/ε^1/3) steps under additional Lipschitz condition.
Bayesian inference engines improve density estimation accuracy and scalability.
problem Constructing accurate and scalable probability density functions.
method Bayesian inference engines (no-U-turn sampling and expectation propagation) with binning strategy.
result Density estimates have excellent comparative performance and scale well to large sample sizes.
L-HNNs improve Bayesian inference by reducing gradient requirements and improving ESS.
problem Efficient Bayesian inference with complex target densities.
method Latent Hamiltonian Neural Networks (L-HNNs) with NUTS, incorporating online error monitoring.
result L-HNNs in NUTS with online error monitoring required 1--2 orders of magnitude fewer numerical gradients and improved ESS by an order of magnitude.
New method trains neural samplers without simulation, but fails due to mode collapse.
problem Training neural samplers without simulation.
method Time-dependent normalizing flow with Langevin preconditioning.
result Langevin preconditioning is crucial for avoiding mode collapse.
Corrected samplers reduce discretization error in discrete flow models without additional computational cost.
problem Discretization error in samplers for discrete flow models.
method Established non-asymptotic error bounds for samplers, proposed time-corrected and location-corrected samplers.
result Location-corrected sampler has lower complexity and better generation quality.
New algorithms sample structured logconcave families with improved efficiency.
problem Sampling structured logconcave families to high accuracy.
method Reduction framework inspired by proximal point methods, combined with restricted Gaussian oracles.
result Improved bounds for sampling structured distributions, matching or surpassing state-of-the-art results.
New method tunes SMC samplers efficiently without high costs.
problem Tuning SMC samplers with unadjusted kernels is challenging.
method Greedy Incremental Divergence Minimization (GIDM) for step size tuning.
result GIDM reduces KL divergence and tunes SMC samplers efficiently.
Gibbs sampler mixes quickly for certain smooth distributions.
problem Drawing samples from log-smooth log-concave distributions.
method Analyzes Gibbs sampler on log-smooth and strongly log-concave distributions.
result Gibbs sampler mixes in O ⋆ ( κ 2 n 7.5 ) O^{\star}(κ^2 n^{7.5}) O ⋆ ( κ 2 n 7.5 ) steps. iDEM generates samples from Boltzmann densities without data.
problem Generating statistically independent samples from unnormalized distributions.
method Iterative algorithm using energy and gradient for diffusion-based sampler training.
result iDEM achieves state-of-the-art performance and trains faster than existing methods.
New sampler improves uniform sampling over convex bodies with fewer queries.
problem Improving uniform sampling over convex bodies with fewer queries.
method Proximal sampler with uniform ergodicity and annealing scheme.
result Converges in Rényi-infinity divergence with O ~ ( d 3 e x t p o l y l o g 1 ε ) \widetilde{\mathcal{O}}(d^3\, ext{polylog} \frac{1}{\varepsilon}) O ( d 3 e x t p o l y l o g ε 1 ) query complexity. We introduce a novel stochastic version of the non-reversible, rejection-free Bouncy Particle Sampler (BPS), a Markov process whose sample trajectories are piecewise linear. The algorithm is based on simulating first arrival times in a doubly stochastic Poisson process using the thinning method, and allows efficient sa…
We propose a novel approach to approximate complex high-dimensional posteriors using neural samplers.
problem Accurately capturing complex, multimodal, and correlated posteriors in high-dimensional spaces.
method Implicit variational inference with neural samplers and local linearisation bounds.
result Our method enables implicit distributions over tens of millions of latent variables, recovering correlations across layers in large Bayesian neural networks.
This work improves the Euler method for masked diffusion models, providing tighter convergence guarantees.
problem Improving the convergence rates of masked diffusion models.
method Developed a direct total-variation (TV) based analysis for the Euler method, relaxing assumptions and improving parameter dependencies.
result Established convergence guarantees for the Euler sampler without requiring surrogate initialization, and provided a tight lower bound.
Stochastic attention learns to retrieve and generate from memory without training.
problem Learning to retrieve and generate from memory efficiently.
method Langevin dynamics on modern Hopfield energy for stochastic attention.
result Stochastic attention can retrieve and generate from memory without training, matching gold standards.
Monte Carlo methods are essential tools for Bayesian inference. Gibbs sampling is a well-known Markov chain Monte Carlo (MCMC) algorithm, extensively used in signal processing, machine learning, and statistics, employed to draw samples from complicated high-dimensional posterior distributions. The key point for the suc…
We present an new sequential Monte Carlo sampler for coalescent based Bayesian hierarchical clustering. Our model is appropriate for modeling non-i.i.d. data and offers a substantial reduction of computational cost when compared to the original sampler without resorting to approximations. We also propose a quadratic co…
Effective implementations of sampling-based probabilistic inference often require manually constructed, model-specific proposals. Inspired by recent progresses in meta-learning for training learning agents that can generalize to unseen environments, we propose a meta-learning approach to building effective and generali…
This article reviews the Author-Topic Model and presents a new non-parametric extension based on the Hierarchical Dirichlet Process. The extension is especially suitable when no prior information about the number of components necessary is available. A blocked Gibbs sampler is described and focus put on staying as clos…
New theory improves diffusion model convergence for generating data.
problem Improving convergence of diffusion models for data generation.
method Developed a non-asymptotic convergence theory for probability flow ODEs.
result Proves d / ε d/\varepsilon d / ε iterations suffice for approximating target distributions. New method simplifies Bayesian analysis for categorical data.
problem Difficulties in scaling GLMs for categorical data due to non-conjugacy or posterior dependencies.
method Defining CB models with binary approximations for tractable inference.
result Fast and scalable inference for thousands of categories, outperforming competitors.
New transport method simplifies cutoff phenomenon for Markov processes.
problem Understanding the cutoff phenomenon for Markov processes.
method A new W-TV transport inequality combined with a parabolic regularization estimate.
result Recovery and extension of previous results on cutoff phenomena.
New samplers improve MCMC efficiency in high dimensions.
problem Efficient sampling in high-dimensional problems.
method Affine invariant ensemble samplers, including derivative-free and derivative-based HMC.
result Affine invariant ensemble HMC outperforms standard HMC in high dimensions.
New theory improves diffusion models' convergence rates.
problem Understanding and optimizing diffusion models for faster data generation.
method Developed non-asymptotic theory for diffusion models with minimal assumptions.
result Established convergence rates for two diffusion models.
We present a new notion of probabilistic duality for random variables involving mixture distributions. Using this notion, we show how to implement a highly-parallelizable Gibbs sampler for weakly coupled discrete pairwise graphical models with strictly positive factors that requires almost no preprocessing and is easy …
New study shows Gaussian samplers struggle with heavy-tailed targets, while stable samplers excel.
problem The difficulty of sampling from heavy-tailed distributions using Gaussian versus stable oracles.
method Comparison of Gaussian and stable oracles for proximal samplers.
result Gaussian samplers have a fundamental barrier for high-accuracy guarantees in heavy-tailed sampling, while stable samplers excel.
SRO optimizes decisions against worst-case sampler induced by generative models.
problem Operational uncertainty shifts from explicit probability law to sampler induced by learned generators.
method SRO optimizes decisions against the worst-case sampler induced by perturbing the learned generator.
result Empirical worst-case objective provides high-probability upper certificate for true population objective.
The Gibbs sampler is a particularly popular Markov chain used for learning and inference problems in Graphical Models (GMs). These tasks are computationally intractable in general, and the Gibbs sampler often suffers from slow mixing. In this paper, we study the Swendsen-Wang dynamics which is a more sophisticated Mark…
Adaptive sampler improves recommendation for implicit feedback data.
problem Challenges in predicting user preferences from implicit feedback data.
method Noisy-label robust learning for adaptive sampler design.
result Significant improvement in recommendation quality on real-world datasets.
Diffusion models simulate molecular dynamics with adjustable accuracy.
problem Simulating molecular dynamics with high accuracy and efficiency.
method Diffusion models as Euler-Maruyama integrators for Langevin dynamics, learning forces from static snapshots.
result Diffusion models generate molecular trajectories with temporal correlations similar to MD simulations.
Discrete diffusion samplers improve sampling from unnormalised densities.
problem Sampling from discrete unnormalised densities efficiently.
method Introduce off-policy training techniques and data-to-energy Schrödinger bridge training for discrete diffusion samplers.
result Improved performance on synthetic and new benchmarks.