New algorithm samples neural network posteriors efficiently.
problem Challenges of sampling multimodal Bayesian posteriors for neural networks.
method Greedy Bayes method using log-concave coupling of posterior and auxiliary random variable.
result Log-concave coupling facilitates efficient sampling of neuron weights.
Enhances SGLD for log-concave posteriors with asynchronous computation.
problem Sampling log-concave posterior distributions efficiently.
method Integrates asynchronous computation into SGLD with delayed gradients.
result Convergence in measure is not significantly affected by delayed gradient information.
This paper optimizes Bayesian estimation for log-concave models using Langevin Monte-Carlo.
problem Optimizing Bayesian estimators for log-concave models with Langevin Monte-Carlo.
method Quantitative statistical bounds and numerical approximation of Gibbs measures.
result Established optimal numerical strategy and its cost for Bayesian posterior mean approximation.
New method improves posterior sampling for complex data models.
problem Sampling from posterior distributions in high-dimensional data.
method Tilted transport technique combining denoising oracle and log-likelihood.
result Boosted posterior is strongly log-concave, facilitating easier sampling.
New method improves sampling for weakly log-concave posteriors.
problem Sampling from weakly log-concave posterior distributions.
method Stochastic Langevin Monte Carlo with over-damped diffusion.
result Simulation horizon is ( d log ( n ) 2 ) ( 1 + r ) 2 (d \log(n)^2)^{(1+r)^2} ( d log ( n ) 2 ) ( 1 + r ) 2 with Poisson subsampling. LaPSRL achieves optimal regret for isoperimetric RL distributions.
problem Designing RL algorithms with sublinear regret for non-log-concave distributions.
method Posterior Sampling (PSRL) and Langevin sampling (LaPSRL) for isoperimetric distributions.
result LaPSRL achieves order-optimal regret and subquadratic complexity.
Hybrid method improves sampling from multimodal distributions.
problem Sampling from multimodal posterior distributions efficiently.
method Jump-Diffusion Langevin Dynamics hybrid with Metropolis.
result Calibrated hybrid method outperforms pure methods.
This work proposes new methods for variational inference using gradient flows on Gaussian measures.
problem Developing algorithmic guarantees for variational inference.
method Proposes principled methods for variational inference using gradient flows on the Bures--Wasserstein space of Gaussian measures.
result Strong theoretical guarantees for log-concave posteriors.
As an important Markov Chain Monte Carlo (MCMC) method, stochastic gradient Langevin dynamics (SGLD) algorithm has achieved great success in Bayesian learning and posterior sampling. However, SGLD typically suffers from slow convergence rate due to its large variance caused by the stochastic gradient. In order to allev…
We consider the problem of transforming samples from one continuous source distribution into samples from another target distribution. We demonstrate with optimal transport theory that when the source distribution can be easily sampled from and the target distribution is log-concave, this can be tractably solved with c…
Improved sampling for high-dimensional posteriors with underdamped Langevin.
problem Scalability issues in high-dimensional problems with approximate Thompson sampling.
method Underdamped Langevin Monte Carlo for accelerated posterior concentration.
result Logarithmic regret improvement from i l d e O ( d ) \mathcal{ ilde O}(d) i l d e O ( d ) to i l d e O ( d ) \mathcal{ ilde O}(\sqrt{d}) i l d e O ( d ) . Method solves Bayesian inverse problems in function space without assuming log-concavity.
problem Bayesian inverse problems in infinite-dimensional nonlinear settings.
method Score-based diffusion models as a prior, Langevin-type MCMC on function spaces.
result Provable convergence bound for posterior sampling, dependent on score approximation.
Combining diffusion models with Langevin dynamics improves posterior sampling efficiency.
problem Sampling from noisy posterior distributions efficiently.
method Annealed Langevin dynamics combined with diffusion models.
result Achieves posterior sampling in polynomial time with a weaker score error bound.
New method accelerates Bayesian imaging using Langevin sampling.
problem Bayesian inference in imaging inverse problems with convex geometry.
method Stochastic relaxed proximal-point iteration targeting posterior distribution.
result Accelerated convergence for κ κ κ -strongly log-concave targets. We study the problem of sampling from the power posterior distribution in Bayesian Gaussian mixture models, a robust version of the classical posterior. This power posterior is known to be non-log-concave and multi-modal, which leads to exponential mixing times for some standard MCMC algorithms. We introduce and study …
Proves log-concavity of cluster algebra coefficients for type A n A_n A n .
problem Log-concavity of cluster algebra coefficients.
method Introduced atomic theta basis and proved log-concavity for type A n A_n A n . result Proved log-concavity of coefficients for cluster algebra variables of type A n A_n A n . Study improves sampling from non-log-concave distributions using Fisher information.
problem Sampling from non-log-concave distributions with high Fisher information guarantees.
method Proximal sampler with RGO implementation, leveraging log-concave sampling results.
result Improved complexity guarantee in relative Fisher information for non-log-concave sampling.
Establishes log-concavity estimates for convex domains' first Dirichlet eigenfunctions.
problem Quantifying the Hessian of log-concave eigenfunctions on convex domains.
method Analyzes log-concavity properties of the first Dirichlet eigenfunction on convex domains.
result Obtains quantitative estimates for the Hessian of log u \log u log u . Log-concavity of eigenfunctions on curved surfaces is proven, leading to fundamental gap estimates.
problem Proving log-concavity of eigenfunctions on curved surfaces.
method Analyzing the Laplacian eigenfunctions on positively curved surfaces.
result Strong log-concavity of the first eigenfunction on positively curved surfaces.
Improved sampling guarantees for weakly log-concave distributions.
problem Sampling from distributions that are not strongly log-concave.
method Proximal sampler with convergence guarantees under weaker assumptions.
result New state-of-the-art sampling guarantees for various target distributions.
We propose a computationally efficient random walk on a convex body which rapidly mixes and closely tracks a time-varying log-concave distribution. We develop general theoretical guarantees on the required number of steps; this number can be calculated on the fly according to the distance from and the shape of the next…
Log-concavity proven for multinomial likelihoods under specific constraints.
problem Log-concavity of multinomial likelihoods under interval censoring constraints.
method Proved log-concavity by showing M-convex subsets of the discrete simplex.
result Likelihood function is completely log-concave.
DE-PSGLD samples from constrained distributions in a decentralized manner.
problem Sampling from log-concave distributions with constraints.
method Decentralized Proximal Stochastic Gradient Langevin Dynamics with proximal regularization.
result DE-PSGLD converges to a regularized Gibbs distribution and maintains posterior concentration.
The Links-Gould polynomial of alternating knots is shown to be log-concave and positive.
problem Verifying the positivity and log-concavity of the Links-Gould polynomial for alternating knots.
method Formulated a conjecture and verified it computationally for all 51.3 million knots with up to 19 crossings.
result All but 544 knots satisfy a stronger log-concavity condition.
Introduces CSLC models to bridge deep generative models and classical algorithms.
problem Mode collapse and memorization issues in deep generative models and restrictive assumptions in classical algorithms.
method Introduces conditionally strongly log-concave (CSLC) models, factorizing data distribution into strongly log-concave conditional distributions.
result Efficient parameter estimation and sampling algorithms with theoretical guarantees for non-log-concave data distributions.
Estimates log-concave densities in graphical models using tent functions.
problem Maximum likelihood estimation of log-concave densities in undirected graphs.
method MLE as product of tent functions corresponding to maximal cliques.
result MLE can be found via convex optimization.
Paper proves super log-concavity of first eigenfunction for certain hyperbolic domains.
problem Proving super log-concavity of first eigenfunction for horo-convex domains in hyperbolic space.
method Analyzes properties of Laplacian eigenfunctions in hyperbolic geometry.
result Optimal proof of super log-concavity for horo-convex domains with constraints.
The paper tackles sampling from Gibbs measures with constrained support, providing a sampling guarantee.
problem Sampling from Gibbs measures with constrained support, especially in the pre-asymptotic regime.
method Analyzing the spectral gap of Langevin dynamics to provide a non-asymptotic sampling guarantee.
result The low-temperature Gibbs distribution concentrates on a neighborhood of its mode in the pre-asymptotic regime.
The paper develops inequalities for log-concave functions and related surface areas.
problem Understanding log-concave functions and their inequalities.
method Establishing new inequalities through f-divergences and functional affine surface areas.
result New inequalities on functional affine surface area and bounds for Kullback-Leibler divergence.
Hamiltonian Monte Carlo (HMC) is a widely deployed method to sample from high-dimensional distributions in Statistics and Machine learning. HMC is known to run very efficiently in practice and its popular second-order "leapfrog" implementation has long been conjectured to run in d 1 / 4 d^{1/4} d 1/4 gradient evaluations. Here we …
Gibbs sampler contracts entropy under strong log-concavity, improving mixing time.
problem Improving the mixing time of Gibbs sampler under strong log-concavity.
method Analyzing Gibbs sampler contraction under strong log-concavity, providing sharp contraction rate.
result Gibbs sampler contracts entropy linearly with condition number and independent of dimension under strong log-concavity.
We construct a compact symplectic manifold with a Hamiltonian circle action for which the Duistermaat-Heckman function is not log-concave.
Log-concave coefficient sequences for two-bridge knots proved.
problem Proving log-concavity of Alexander polynomial coefficient sequences for alternating knots.
method Introducing a polynomial Δ ( t ) Δ(t) Δ ( t ) associated to Christoffel words and proving its log-concavity. result Strong Fox conjecture for two-bridge knots proved.
CAVI converges for log-concave measures via optimal transport.
problem Finding the closest product measure to a log-concave measure via CAVI.
method Adapting coordinate descent techniques from Euclidean space to optimal transport for log-concave densities.
result Proves convergence of CAVI for log-concave densities and provides rates of convergence under additional conditions.
Zigzag sampling algorithm efficiently samples from strongly log-concave distributions with low computational cost.
problem Sampling from strongly log-concave distributions efficiently and with low computational complexity.
method Zigzag sampling algorithm with warm start assumption, focusing on gradient evaluations.
result Achieves ε error in chi-square divergence with computational cost of O(κ²d^(1/2)(log(1/ε))^(3/2)) gradient evaluations.
This work improves bounds on Bayesian coreset quality.
problem Limitations of existing theoretical analysis of Bayesian coresets.
method Develops general upper and lower bounds on KL divergence.
result Demonstrates flexibility of new theoretical bounds in various models.
New lower bounds for sampling from log-concave distributions in higher dimensions.
problem Proving lower bounds for sampling from log-concave distributions in higher dimensions.
method Multiscale construction inspired by geometric measure theory and reduction to block Krylov algorithms.
result Query lower bounds for sampling from log-concave distributions in higher dimensions are established.
Guarantees convergence for black-box variational inference without modifications.
problem Convergence guarantees for black-box variational inference.
method Analysis of log-smooth posterior densities, location-scale variational family, and convergence rates of algorithm design choices.
result Proximal stochastic gradient descent fixes suboptimal convergence rates and achieves strongest known guarantees.
A key task in Bayesian machine learning is sampling from distributions that are only specified up to a partition function (i.e., constant of proportionality). One prevalent example of this is sampling posteriors in parametric distributions, such as latent-variable generative models. However sampling (even very approxim…
FA-LD algorithm improves uncertainty quantification and mean predictions in federated learning.
problem Uncertainty quantification and mean predictions in federated learning with distributed clients.
method FA-LD algorithm for strongly log-concave distributions with non-i.i.d data, considering general models.
result The FA-LD algorithm provides theoretical guarantees for convergence and optimal noise injection.
Study Langevin Monte Carlo for sampling non-log-concave distributions.
problem Sampling from non-log-concave distributions, especially Gaussian mixtures.
method Discretizations of overdamped Langevin diffusions.
result Numerical simulations compare Langevin Monte Carlo algorithms' performance.
Karshon constructed the first counterexample to the log-concavity conjecture for the Duistermaat-Heckman measure: a Hamiltonian six manifold whose fixed points set is the disjoint union of two copies of T 4 T^4 T 4 . In this article, for any closed symplectic four manifold N N N with b + b+ b + greater than 1, we show that there is a…
It is well known that Markov chain Monte Carlo (MCMC) methods scale poorly with dataset size. A popular class of methods for solving this issue is stochastic gradient MCMC. These methods use a noisy estimate of the gradient of the log posterior, which reduces the per iteration computational cost of the algorithm. Despi…
Study minimax risk of score estimation for log-concave distributions.
problem Minimizing risk in score estimation for log-concave distributions.
method Developed subclasses of log-concave densities and constructed a locally adaptive, multiscale estimator.
result Established minimax rates for score estimation over specific subclasses of log-concave densities.
RHMC accelerates sampling from log-concave distributions.
problem Sampling from log-concave probability distributions efficiently.
method RHMC uses simulated Hamiltonian dynamics with random integration times.
result RHMC converges exponentially fast in KL divergence for log-concave distributions.
Improved sampling for diffusion models and log-concave distributions.
problem Efficient sampling for diffusion models and log-concave distributions.
method Algorithms for sampling with δ δ δ -error in p o l y l o g ( 1 / δ ) \mathrm{polylog}(1/δ) polylog ( 1/ δ ) steps using accurate score estimates. result Exponential improvement in complexity over previous results.
The main results are two characterisations of log-concave densities in terms of the collection of lift zonoids corresponding to a peacock. These notions are recalled and connected to arbitrage-free asset pricing in financial mathematics.
This work studies the location estimation problem for a mixture of two rotation invariant log-concave densities. We demonstrate that Least Squares EM, a variant of the EM algorithm, converges to the true location parameter from a randomly initialized point. We establish the explicit convergence rates and sample complex…