arXiv research

The paper analyzes the InfoNCE loss under different temperature schedules using Langevin dynamics.

Study improves sampling from complex distributions using annealed Langevin Monte Carlo.

Study of diffusion annealed Langevin dynamics for generative models.

DALMC provides non-asymptotic error bounds for generative models.

Paper analyzes Annealed Langevin Dynamics for multimodal sampling stability.

Paper analyzes Langevin dynamics for multimodal Gaussian mixtures, controlling errors across dimensions.

CoolMomentum combines momentum and Simulated Annealing for deep learning optimization.

Annealed Langevin dynamics improves sampling from composite scores in SBI.

Score matching method improves image generation quality.

Combining diffusion models with Langevin dynamics improves posterior sampling efficiency.

Proposes efficient sampling methods for solving linear inverse problems.

RLD improves combinatorial optimization by avoiding local minima.

AIS uses a suboptimal extended target distribution, which this paper improves using SGM.

Gradient-descent-based algorithms and their stochastic versions have widespread applications in machine learning and statistical inference. In this work we perform an analytic study of the performances of one of them, the Langevin algorithm, in the context of noisy high-dimensional inference. We employ the Langevin alg…

Algorithm estimates graph structure with prior information and Langevin diffusion.

New method RECAST improves uncertainty calibration in neural networks.

The paper analyzes convergence of Langevin dynamics with time-dependent metrics.

Novel method for nonlinear data assimilation using Langevin sampling.

Flow Annealing Posterior Sampling unifies stochastic-process regression and PDE inverse problems.

Kernel SVGD improves high-dimensional inference with noise adaptation.

New method samples from posterior distributions efficiently.

We consider the problem of minimizing a convex objective function $F$ when one can only evaluate its noisy approximation $\hat{F}$. Unless one assumes some structure on the noise, $\hat{F}$ may be an arbitrary nonconvex function, making the task of minimizing $F$ intractable. To overcome this, prior work has often focu…

New method combines QQA and gradient-based sampling for combinatorial optimization.

DAIS improves AIS for differentiable marginal likelihood estimation.

In this paper, we propose a novel uniform generalization bound on the time and inverse temperature for stochastic gradient Langevin dynamics (SGLD) in a non-convex setting. While previous works derive their generalization bounds by uniform stability, we use Rademacher complexity to make our generalization bound indepen…

Bayesian approach uses generative models as priors for better source separation.

We introduce a new generative model where samples are produced via Langevin dynamics using gradients of the data distribution estimated with score matching. Because gradients can be ill-defined and hard to estimate when the data resides on low-dimensional manifolds, we perturb the data with different levels of Gaussian…

Combines SMC and diffusion-based samplers for improved sampling performance.

Paper extends MLFD to signed measures via bilevel approach.

Improves generative models by using heavy-tailed noise in score matching.

New algorithms estimate normalizing constants for log-concave distributions efficiently.

Reverse annealing boosts quantum matrix factorization performance.

A new sampling method estimates scores without training or nested MCMC.

We empirically evaluate a stochastic annealing strategy for Bayesian posterior optimization with variational inference. Variational inference is a deterministic approach to approximate posterior inference in Bayesian models in which a typically non-convex objective function is locally optimized over the parameters of t…

New method uses reinforcement learning to improve Simulated Annealing.

This work extends diffusion models to function space for better generative modeling.

aMCL uses annealing to improve hypothesis diversity in ambiguous tasks.

This paper presents studies on a deterministic annealing algorithm based on quantum annealing for variational Bayes (QAVB) inference, which can be seen as an extension of the simulated annealing for variational Bayes (SAVB) inference. QAVB is as easy as SAVB to implement. Experiments revealed QAVB finds a better local …

Quantum annealers aim at solving non-convex optimization problems by exploiting cooperative tunneling effects to escape local minima. The underlying idea consists in designing a classical energy function whose ground states are the sought optimal solutions of the original optimization problem and add a controllable qua…

Improves diffusion model performance and efficiency through classical search.

Mathematical analysis shows annealing prevents mode collapse in Gaussian mixtures.

Proposes a method to improve SLMC for multimodal distributions.

We introduce a novel framework for adversarial training where the target distribution is annealed between the uniform distribution and the data distribution. We posited a conjecture that learning under continuous annealing in the nonparametric regime is stable irrespective of the divergence measures in the objective fu…

Develops diffusion samplers for target distributions with efficient score and density estimates.

Simulated annealing improves candidate optimization for multi-objective Bayesian optimization.

New analysis of annealing paths in sampling and estimation.

AdaAnn optimizes annealing for efficient probability density approximation.

CRAFT improves on existing methods for sampling complex distributions.