arXiv research

A new method combines Gaussian graphical models for better distributed Gaussian process predictions.

This work shows that Gaussian is the only prior for optimal linear estimation in $L^1$ loss.

GGMPs improve non-Gaussian conditional density estimation.

Conditional diffusion models can approximate target distributions well with Gaussian-mixture reverse kernels.

This paper considers inference over distributed linear Gaussian models using factor graphs and Gaussian belief propagation (BP). The distributed inference algorithm involves only local computation of the information matrix and of the mean vector, and message passing between neighbors. Under broad conditions, it is show…

The paper shows how to infer conditional independence from non-Gaussian data.

This work introduces efficient sampling methods for Gaussian processes by focusing on pathwise conditioning.

Paper explores exact recovery of communities in weighted graphs using Gaussian and exponential distributions.

Study selective classification with halfspaces, achieving error bounds under Gaussian distributions.

FlowSDR learns a low-dimensional projection preserving the response's conditional distribution.

TSFlow uses Gaussian processes to match priors for better time series forecasting.

We establish large deviation principles for convolutional neural networks.

GP-ConvCNP improves NP models for time series data by adding Gaussian Process.

Deep neural networks converge to Gaussian mixtures as layer width increases.

HTFM improves mode coverage and tail-statistic recovery for heavy-tailed data.

Ens-CGP synthesizes ensemble-based inference with Gaussian processes.

Conditional Density Estimation (CDE) models deal with estimating conditional distributions. The conditions imposed on the distribution are the inputs of the model. CDE is a challenging task as there is a fundamental trade-off between model complexity, representational capacity and overfitting. In this work, we propose …

Efficiently learns Gaussian tree models with near-optimal sample complexity.

Gaussians as noise in NCE lead to exponentially bad conditioning, hindering its efficiency.

Elliptical processes generalize Gaussian and Student-t models with fat tails and computational efficiency.

New SQ lower bounds for NGCA without requiring chi-squared condition.

MD-CGAN models forecast time series with probabilistic posterior distributions.

Knowing when a graphical model is perfect to a distribution is essential in order to relate separation in the graph to conditional independence in the distribution, and this is particularly important when performing inference from data. When the model is perfect, there is a one-to-one correspondence between conditional…

Algorithm learns non-Gaussian graphical models via Hessian scores and triangular transport.

CGDL improves open set recognition by learning conditional Gaussian distributions.

Researchers disrupt Gaussian model inference to test adversarial attacks.

HMC improves Gaussian sampling efficiency with long, random steps.

The paper proves the convergence of Q-value for Gaussian rewards.

Bayesian inference for wide neural networks using Edgeworth expansion.

A new copula model for multi-attribute data using optimal transport.

This paper addresses the problem of scalable optimization for L1-regularized conditional Gaussian graphical models. Conditional Gaussian graphical models generalize the well-known Gaussian graphical models to conditional distributions to model the output network influenced by conditioning input variables. While highly …

New sampling-based approach for filtering problems using multiplicative Gaussian functions.

Ancestral graph models, introduced by Richardson and Spirtes (2002), generalize both Markov random fields and Bayesian networks to a class of graphs with a global Markov property that is closed under conditioning and marginalization. By design, ancestral graphs encode precisely the conditional independence structures t…

A robust Gaussian process model using Huber likelihood for outlier resistance.

Paper improves SDR estimation speed and conditions.

Bayesian approach for solving systems of linear PDEs with boundary conditions.

The analysis of observed conditional distributions of both lagged and simultaneous intraday price increments of a basket of stocks reveals phenomena of dependence - induced volatility smile and kurtosis reduction. A model based on multivariate t-Student distribution shows that the observed effects are caused by colelct…

Paper solves NGCA for discrete distributions using LLL method.

A new model forecasts Value-at-Risk using NIG distribution and dynamic scores.

Bayesian approach approximates probability functions of Gaussian mixtures.

This paper proposes a new method for conditional sampling using optimal transport.

We consider the problem of sequential learning from categorical observations bounded in [0,1]. We establish an ordering between the Dirichlet posterior over categorical outcomes and a Gaussian posterior under observations with N(0,1) noise. We establish that, conditioned upon identical data with at least two observatio…

New method aggregates Gaussian experts by detecting conditional independence violations.

Gaussianization flows transform any random vector into a Gaussian, enabling efficient computation and sample generation.

A new method for learning conditional distributions using ODEs and neural networks.

Improving Bayesian Optimization via Training-Aware Conditional Diffusion Models

New EM algorithm for mixtures of elliptical distributions handles missing data and outliers.

We study learning problems in which the conditional distribution of the output given the input varies as a function of additional task variables. In varying-coefficient models with Gaussian process priors, a Gaussian process generates the functional relationship between the task variables and the parameters of this con…