arXiv subjects

Wing Hung Wong

Publications and source records attributed to Wing Hung Wong.

At least 19 recordsLinked to original sources

arXiv2026

An AI-powered Bayesian Generative Modeling Approach for Arbitrary Conditional Inference

Modern data analysis increasingly requires flexible conditional inference P(X_B | X_A) where (X_A, X_B) is an arbitrary partition of observed variable X. Existing approaches are either restricted to a fixed conditioning structure or depend strongly on the distribution of conditioning masks during training. To address these limitations, we introduce Bayesian generative modeling (BGM), a unified framework for arbitrary conditional inference. BGM learns a generative model of X via a stochastic iterative Bayesian updating algorithm in which model parameters and latent variables are updated until convergence. Once trained, any conditional distribution can be obtained without retraining. Empirically, BGM achieves superior predictive performance with posterior predictive intervals, demonstrating that a single learned model can serve as a universal engine for conditional prediction with principled uncertainty quantification. We provide theoretical guarantees for convergence of the stochastic iterative algorithm, statistical consistency, and conditional risk bounds. The proposed BGM framework leverages modern AI to capture complex relationships among variables while adhering to Bayesian principles, offering a promising approach for a wide range of applications in modern data science. Code for BGM is available at https://github.com/liuq-lab/bayesgm. Document of BGM is available at https://bayesgm.readthedocs.io.

stat.ML↗

arXiv2026

An AI-powered Bayesian generative modeling approach for causal inference in observational studies

Causal inference in observational studies with high-dimensional covariates presents significant challenges. We introduce CausalBGM, an AI-powered Bayesian generative modeling approach that captures the causal relationship among covariates, treatment, and outcome. The core innovation is to estimate the individual treatment effect (ITE) by learning the individual-specific distribution of a low-dimensional latent feature set (e.g., latent confounders) that drives changes in both treatment and outcome. This individualized posterior representation yields estimates of the individual treatment effect (ITE) together with well-calibrated posterior intervals while mitigating confounding effect. CausalBGM is fitted through an iterative algorithm to update the model parameters and the latent features until convergence. This framework leverages the power of AI to capture complex dependencies among variables while adhering to the Bayesian principles. Extensive experiments demonstrate that CausalBGM consistently outperforms state-of-the-art methods, particularly in scenarios with high-dimensional covariates and large-scale datasets. By addressing key limitations of existing methods, CausalBGM emerges as a robust and promising framework for advancing causal inference in a wide range of modern applications. The code for CausalBGM is available at https://github.com/liuq-lab/bayesgm. The document for using CausalBGM is available at https://bayesgm.readthedocs.io.

stat.ML↗

arXiv2025

Likelihood-Free Adaptive Bayesian Inference via Nonparametric Distribution Matching

When the likelihood is analytically unavailable and computationally intractable, approximate Bayesian computation (ABC) has emerged as a widely used methodology for approximate posterior inference; however, it suffers from severe computational inefficiency in high-dimensional settings or under diffuse priors. To overcome these limitations, we propose Adaptive Bayesian Inference (ABI), a framework that bypasses traditional data-space discrepancies and instead compares distributions directly in posterior space through nonparametric distribution matching. By leveraging a novel Marginally-augmented Sliced Wasserstein (MSW) distance on posterior measures and exploiting its quantile representation, ABI transforms the challenging problem of measuring divergence between posterior distributions into a tractable sequence of one-dimensional conditional quantile regression tasks. Moreover, we introduce a new adaptive rejection sampling scheme that iteratively refines the posterior approximation by updating the proposal distribution via generative density estimation. Theoretically, we establish parametric convergence rates for the trimmed MSW distance and prove that the ABI posterior converges to the true posterior as the tolerance threshold vanishes. Through extensive empirical evaluation, we demonstrate that ABI significantly outperforms data-based Wasserstein ABC, summary-based ABC, and state-of-the-art likelihood-free simulators, especially in high-dimensional or dependent observation regimes.

stat.ME↗

arXiv2025

Efficient Generative Modeling via Penalized Optimal Transport Network

The generation of synthetic data with distributions that faithfully emulate the underlying data-generating mechanism holds paramount significance. Wasserstein Generative Adversarial Networks (WGANs) have emerged as a prominent tool for this task; however, due to the delicate equilibrium of the minimax formulation and the instability of Wasserstein distance in high dimensions, WGAN often manifests the pathological phenomenon of mode collapse. This results in generated samples that converge to a restricted set of outputs and fail to adequately capture the tail behaviors of the true distribution. Such limitations can lead to serious downstream consequences. To this end, we propose the Penalized Optimal Transport Network (POTNet), a versatile deep generative model based on the marginally-penalized Wasserstein (MPW) distance. Through the MPW distance, POTNet effectively leverages low-dimensional marginal information to guide the overall alignment of joint distributions. Furthermore, our primal-based framework enables direct evaluation of the MPW distance, thus eliminating the need for a critic network. This formulation circumvents training instabilities inherent in adversarial approaches and avoids the need for extensive parameter tuning. We derive a non-asymptotic bound on the generalization error of the MPW loss and establish convergence rates of the generative distribution learned by POTNet. Our theoretical analysis together with extensive empirical evaluations demonstrate the superior performance of POTNet in accurately capturing underlying data structures, including their tail behaviors and minor modalities. Moreover, our model achieves orders of magnitude speedup during the sampling stage compared to state-of-the-art alternatives, which enables computationally efficient large-scale synthetic data generation.

stat.ML↗

arXiv2023

CausalEGM: a general causal inference framework by encoding generative modeling

Although understanding and characterizing causal effects have become essential in observational studies, it is challenging when the confounders are high-dimensional. In this article, we develop a general framework $\textit{CausalEGM}$ for estimating causal effects by encoding generative modeling, which can be applied in both binary and continuous treatment settings. Under the potential outcome framework with unconfoundedness, we establish a bidirectional transformation between the high-dimensional confounders space and a low-dimensional latent space where the density is known (e.g., multivariate normal distribution). Through this, CausalEGM simultaneously decouples the dependencies of confounders on both treatment and outcome and maps the confounders to the low-dimensional latent space. By conditioning on the low-dimensional latent features, CausalEGM can estimate the causal effect for each individual or the average causal effect within a population. Our theoretical analysis shows that the excess risk for CausalEGM can be bounded through empirical process theory. Under an assumption on encoder-decoder networks, the consistency of the estimate can be guaranteed. In a series of experiments, CausalEGM demonstrates superior performance over existing methods for both binary and continuous treatments. Specifically, we find CausalEGM to be substantially more powerful than competing methods in the presence of large sample sizes and high dimensional confounders. The software of CausalEGM is freely available at https://github.com/SUwonglab/CausalEGM.

stat.ML↗

arXiv2021

An integral equation for the identification of causal effects in nonlinear models

When the causal relationship between X and Y is specified by a structural equation, the causal effect of X on Y is the expected rate of change of Y with respect to changes in X, when all other variables are kept fixed. This causal effect is not identifiable from the distribution of (X,Y). We give conditions under which this causal effect is identified as the solution of an integral equation based on the distributions of (X,Z) and (Y,Z), where Z is an instrumental variable.

math.ST↗

arXiv2021

A calculus for causal inference with instrumental variables

Under a general structural equation framework for causal inference, we provide a definition of the causal effect of a variable X on another variable Y, and propose an approach to estimate this causal effect via the use of instrumental variables.

math.ST↗

arXiv2020

Roundtrip: A Deep Generative Neural Density Estimator

Density estimation is a fundamental problem in both statistics and machine learning. In this study, we proposed Roundtrip as a general-purpose neural density estimator based on deep generative models. Roundtrip retains the generative power of generative adversarial networks (GANs) but also provides estimates of density values. Unlike previous neural density estimators that put stringent conditions on the transformation from the latent space to the data space, Roundtrip enables the use of much more general mappings. In a series of experiments, Roundtrip achieves state-of-the-art performance in a diverse range of density estimation tasks.

cs.LG↗

arXiv2018

Detecting strong signals in gene perturbation experiments: An adaptive approach with power guarantee and FDR control

The perturbation of a transcription factor should affect the expression levels of its direct targets. However, not all genes showing changes in expression are direct targets. To increase the chance of detecting direct targets, we propose a modified two-group model where the null group corresponds to genes which are not direct targets, but can have small non-zero effects. We model the behaviour of genes from the null set by a Gaussian distribution with unknown variance $τ^2$, and we discuss and compare three methods which adaptively estimate $τ^2$ from the data: the iterated empirical Bayes estimator, the truncated MLE and the central moment matching estimator. We conduct a detailed analysis of the properties of the iterated EB estimate which has the best performance in the simulations. In particular, we provide theoretical guarantee of its good performance under mild conditions. We provide simulations comparing the new modeling approach with existing methods, and the new approach shows more stable and better performance under different situations. We also apply it to a real data set from gene knock-down experiments and obtained better results compared with the original two-group model testing for non-zero effects.

stat.ME↗

arXiv2018

Density Estimation via Discrepancy Based Adaptive Sequential Partition

Given $iid$ observations from an unknown absolute continuous distribution defined on some domain $Ω$, we propose a nonparametric method to learn a piecewise constant function to approximate the underlying probability density function. Our density estimate is a piecewise constant function defined on a binary partition of $Ω$. The key ingredient of the algorithm is to use discrepancy, a concept originates from Quasi Monte Carlo analysis, to control the partition process. The resulting algorithm is simple, efficient, and has a provable convergence rate. We empirically demonstrate its efficiency as a density estimation method. We present its applications on a wide range of tasks, including finding good initializations for k-means.

stat.ML↗

arXiv2016

co-BPM: a Bayesian Model for Divergence Estimation

Divergence is not only an important mathematical concept in information theory, but also applied to machine learning problems such as low-dimensional embedding, manifold learning, clustering, classification, and anomaly detection. We proposed a bayesian model---co-BPM---to characterize the discrepancy of two sample sets, i.e., to estimate the divergence of their underlying distributions. In order to avoid the pitfalls of plug-in methods that estimate each density independently, our bayesian model attempts to learn a coupled binary partition of the sample space that best captures the landscapes of both distributions, then make direct inference on their divergences. The prior is constructed by leveraging the sequential buildup of the coupled binary partitions and the posterior is sampled via our specialized MCMC. Our model provides a unified way to estimate various types of divergences and enjoys convincing accuracy. We demonstrate its effectiveness through simulations, comparisons with the \emph{state-of-the-art} and a real data example.

stat.CO↗

arXiv2016

Discovering and Visualizing Hierarchy in Multivariate Data

How to extract useful insights from data is always a challenge, especially if the data is multidimensional. Often, the data can be organized according to certain hierarchical structure that are stemmed either from data collection process or from the information and phenomena carried by the data itself. The current study attempts to discover and visualize these underlying hierarchies. By regarding each observation in the data as a draw from a (hypothetical) multidimensional joint density, our first goal is to approximate this unknown density with a piecewise constant function via binary partition, our non-parametric approach makes no assumptions on the form of the density. Given the piecewise constant density function and its corresponding binary partition, our second goal is to construct a connected graph and build up a tree representation of the data by level sets. To demonstrate that our method is a general data mining and visualization tool which can provide "multi-resolution" summaries and reveal different levels of information of the data, we apply it to two real data sets from Flow Cytometry and Social Network.

stat.AP↗

arXiv2016

Learning a nonlinear dynamical system model of gene regulation: A perturbed steady-state approach

Biological structure and function depend on complex regulatory interactions between many genes. A wealth of gene expression data is available from high-throughput genome-wide measurement technologies, but effective gene regulatory network inference methods are still needed. Model-based methods founded on quantitative descriptions of gene regulation are among the most promising, but many such methods rely on simple, local models or on ad hoc inference approaches lacking experimental interpretability. We propose an experimental design and develop an associated statistical method for inferring a gene network by learning a standard quantitative, interpretable, predictive, biophysics-based ordinary differential equation model of gene regulation. We fit the model parameters using gene expression measurements from perturbed steady-states of the system, like those following overexpression or knockdown experiments. Although the original model is nonlinear, our design allows us to transform it into a convex optimization problem by restricting attention to steady-states and using the lasso for parameter selection. Here, we describe the model and inference algorithm and apply them to a synthetic six-gene system, demonstrating that the model is detailed and flexible enough to account for activation and repression as well as synergistic and self-regulation, and the algorithm can efficiently and accurately recover the parameters used to generate the data.

q-bio.MN↗

arXiv2015

Multivariate Density Estimation via Adaptive Partitioning (I): Sieve MLE

We study a non-parametric approach to multivariate density estimation. The estimators are piecewise constant density functions supported by binary partitions. The partition of the sample space is learned by maximizing the likelihood of the corresponding histogram on that partition. We analyze the convergence rate of the sieve maximum likelihood estimator, and reach a conclusion that for a relatively rich class of density functions the rate does not directly depend on the dimension. This suggests that, under certain conditions, this method is immune to the curse of dimensionality, in the sense that it is possible to get close to the parametric rate even in high dimensions. We also apply this method to several special cases, and calculate the explicit convergence rates respectively.

math.ST↗

arXiv2015

Multivariate Density Estimation via Adaptive Partitioning (II): Posterior Concentration

In this paper, we study a class of non-parametric density estimators under Bayesian settings. The estimators are piecewise constant functions on binary partitions. We analyze the concentration rate of the posterior distribution under a suitable prior, and demonstrate that the rate does not directly depend on the dimension of the problem. This paper can be viewed as an extension of a parallel work where the convergence rate of a related sieve MLE was established. Compared to the sieve MLE, the main advantage of the Bayesian method is that it can adapt to the unknown complexity of the true density function, thus achieving the optimal convergence rate without artificial conditions on the density.

math.ST↗

arXiv2013

Computational Aspects of Optional Pólya Tree

Optional Pólya Tree (OPT) is a flexible non-parametric Bayesian model for density estimation. Despite its merits, the computation for OPT inference is challenging. In this paper we present time complexity analysis for OPT inference and propose two algorithmic improvements. The first improvement, named Limited-Lookahead Optional Pólya Tree (LL-OPT), aims at greatly accelerate the computation for OPT inference. The second improvement modifies the output of OPT or LL-OPT and produces a continuous piecewise linear density estimate. We demonstrate the performance of these two improvements using simulations.

stat.CO↗

arXiv2011

Statistical Modeling of RNA-Seq Data

Recently, ultra high-throughput sequencing of RNA (RNA-Seq) has been developed as an approach for analysis of gene expression. By obtaining tens or even hundreds of millions of reads of transcribed sequences, an RNA-Seq experiment can offer a comprehensive survey of the population of genes (transcripts) in any sample of interest. This paper introduces a statistical model for estimating isoform abundance from RNA-Seq data and is flexible enough to accommodate both single end and paired end RNA-Seq data and sampling bias along the length of the transcript. Based on the derivation of minimal sufficient statistics for the model, a computationally feasible implementation of the maximum likelihood estimator of the model is provided. Further, it is shown that using paired end RNA-Seq provides more accurate isoform abundance estimates than single end sequencing at fixed sequencing depth. Simulation studies are also given.

stat.ME↗

arXiv2009

Reconstructing the energy landscape of a distribution from Monte Carlo samples

Defining the energy function as the negative logarithm of the density, we explore the energy landscape of a distribution via the tree of sublevel sets of its energy. This tree represents the hierarchy among the connected components of the sublevel sets. We propose ways to annotate the tree so that it provides information on both topological and statistical aspects of the distribution, such as the local energy minima (local modes), their local domains and volumes, and the barriers between them. We develop a computational method to estimate the tree and reconstruct the energy landscape from Monte Carlo samples simulated at a wide energy range of a distribution. This method can be applied to any arbitrary distribution on a space with defined connectedness. We test the method on multimodal distributions and posterior distributions to show that our estimated trees are accurate compared to theoretical values. When used to perform Bayesian inference of DNA sequence segmentation, this approach reveals much more information than the standard approach based on marginal posterior distributions.

stat.AP↗