arXiv research

TMDA aligns subdomain data distribution discrepancies across domains using manifold representations.

Stein discrepancy improves UDA performance in low-data scenarios.

The performance of standard learning procedures has been observed to differ widely across groups. Recent studies usually attribute this loss discrepancy to an information deficiency for one group (e.g., one group has less data). In this work, we point to a more subtle source of loss discrepancy---feature noise. Our mai…

Two methods using low-discrepancy points improve data compression for neural networks.

First principles modeling of physical systems has led to significant technological advances across all branches of science. For nonlinear systems, however, small modeling errors can lead to significant deviations from the true, measured behavior. Even in mechanical systems, where the equations are assumed to be well-kn…

Paper proposes kernelized Stein tests for time-to-event data with censoring.

A new method for kernel tests without data splitting increases power.

We present a new method for evaluating and training unnormalized density models. Our approach only requires access to the gradient of the unnormalized model's log-density. We estimate the Stein discrepancy between the data density $p(x)$ and the model density $q(x)$ defined by a vector function of the data. We paramete…

Framework identifies discrepancies in physics models, improving sensor accuracy.

Bayesian data selection framework ensures fairness in machine learning models.

Unsupervised domain adaptation is the problem setting where data generating distributions in the source and target domains are different, and labels in the target domain are unavailable. One important question in unsupervised domain adaptation is how to measure the difference between the source and target domains. A pr…

MMD test detects adversarial attacks by addressing kernel limitations and non-independence issues.

A new framework improves kernel Stein discrepancy tests for validating distributions.

Optimizes kernel discrepancies by selecting subsets efficiently.

New deep learning method outperforms random training data.

New discrepancy function compares discrete probability measures considering space geometry.

This paper introduces localized discrepancy theories for unsupervised domain adaptation.

Proposes an efficient lower bound for Gromov-Wasserstein discrepancy.

A new ensemble filter uses transport maps and MMD optimization for high-dimensional data assimilation.

The article introduces practical estimators for kernel discrepancies.

Generative adversarial networks (GANs) generate data based on minimizing a divergence between two distributions. The choice of that divergence is therefore critical. We argue that the divergence must take into account the hypothesis set and the loss function used in a subsequent learning task, where the data generated …

Study finds discrepancies in model explanations for clean vs. adversarial inputs.

Private algorithms adapt from public to private domains with minimal labeled data.

MPMC generates low-discrepancy points using graph neural networks.

Sliced kernelized Stein discrepancy improves goodness-of-fit tests and model learning in high dimensions.

A new method trains discrete EBMs without sampling.

Proposes robust ABC method for outlier detection.

Much of machine learning relies on comparing distributions with discrepancy measures. Stein's method creates discrepancy measures between two distributions that require only the unnormalized density of one and samples from the other. Stein discrepancies can be combined with kernels to define kernelized Stein discrepanc…

This work investigates how GCNs should handle local structure discrepancies in testing nodes.

Study shows the corrected Akaike criterion is inadmissible for estimating Kullback-Leibler discrepancy.

A new embedding method extracts dataset-scale metric distribution into vectorial representation for graph data.

Secure SMMD enables data privacy in federated learning.

Fourier representation improves KSD for infinite-dimensional data.

Study on kernel tests for high-dimensional data, focusing on MMD and CLT.

This paper defines the notion of class discrepancy for families of functions. It shows that low discrepancy classes admit small offline and streaming coresets. We provide general techniques for bounding the class discrepancy of machine learning problems. As corollaries of the general technique we bound the discrepancy …

Expands Bayesian experiment design framework to account for model discrepancies.

Semi-parametric framework for nonlinear system identification

A new test statistic measures discrepancy between conditional distributions.

Improved kernel Stein discrepancy for large-scale data.

New partition designs reduce star discrepancy in high-dimensional sampling.

We propose a novel fused Gromov-Wasserstein alignment method to jointly learn the Hawkes processes in different event spaces, and align their event types. Given two Hawkes processes, we use fused Gromov-Wasserstein discrepancy to measure their dissimilarity, which considers both the Wasserstein discrepancy based on the…

New model learns better policies from expert demonstrations with higher efficiency.

A new measure scales MMD to assess distribution closeness.

This work proposes a new method to match distributions across different spaces using cycle-consistent maps.

New algorithm improves knowledge transfer in dynamic decision-making.

A new method for density estimation using mixture discrepancy and moments.

Validates composite systems using discrepancy propagation.

Stochastic Stein Discrepancies improve inference efficiency.