arXiv subjects

Hongtu Zhu

Publications and source records attributed to Hongtu Zhu.

At least 91 records · Page 5Linked to original sources

arXiv2021

An Online Sequential Test for Qualitative Treatment Effects

Tech companies (e.g., Google or Facebook) often use randomized online experiments and/or A/B testing primarily based on the average treatment effects to compare their new product with an old one. However, it is also critically important to detect qualitative treatment effects such that the new one may significantly outperform the existing one only under some specific circumstances. The aim of this paper is to develop a powerful testing procedure to efficiently detect such qualitative treatment effects. We propose a scalable online updating algorithm to implement our test procedure. It has three novelties including adaptive randomization, sequential monitoring, and online updating with guaranteed type-I error control. We also thoroughly examine the theoretical properties of our testing procedure including the limiting distribution of test statistics and the justification of an efficient bootstrap method. Extensive empirical studies are conducted to examine the finite sample performance of our test procedure.

stat.ME↗

arXiv2021

Real-world Ride-hailing Vehicle Repositioning using Deep Reinforcement Learning

We present a new practical framework based on deep reinforcement learning and decision-time planning for real-world vehicle repositioning on ride-hailing (a type of mobility-on-demand, MoD) platforms. Our approach learns the spatiotemporal state-value function using a batch training algorithm with deep value networks. The optimal repositioning action is generated on-demand through value-based policy search, which combines planning and bootstrapping with the value networks. For the large-fleet problems, we develop several algorithmic features that we incorporate into our framework and that we demonstrate to induce coordination among the algorithmically-guided vehicles. We benchmark our algorithm with baselines in a ride-hailing simulation environment to demonstrate its superiority in improving income efficiency meausred by income-per-hour. We have also designed and run a real-world experiment program with regular drivers on a major ride-hailing platform. We have observed significantly positive results on key metrics comparing our method with experienced drivers who performed idle-time repositioning based on their own expertise.

cs.LG↗

arXiv2021

A Deep Value-network Based Approach for Multi-Driver Order Dispatching

Recent works on ride-sharing order dispatching have highlighted the importance of taking into account both the spatial and temporal dynamics in the dispatching process for improving the transportation system efficiency. At the same time, deep reinforcement learning has advanced to the point where it achieves superhuman performance in a number of fields. In this work, we propose a deep reinforcement learning based solution for order dispatching and we conduct large scale online A/B tests on DiDi's ride-dispatching platform to show that the proposed method achieves significant improvement on both total driver income and user experience related metrics. In particular, we model the ride dispatching problem as a Semi Markov Decision Process to account for the temporal aspect of the dispatching actions. To improve the stability of the value iteration with nonlinear function approximators like neural networks, we propose Cerebellar Value Networks (CVNet) with a novel distributed state representation layer. We further derive a regularized policy evaluation scheme for CVNet that penalizes large Lipschitz constant of the value network for additional robustness against adversarial perturbation and noises. Finally, we adapt various transfer learning methods to CVNet for increased learning adaptability and efficiency across multiple cities. We conduct extensive offline simulations based on real dispatching data as well as online AB tests through the DiDi's platform. Results show that CVNet consistently outperforms other recently proposed dispatching methods. We finally show that the performance can be further improved through the efficient use of transfer learning.

cs.LG↗

arXiv2021

Value Function is All You Need: A Unified Learning Framework for Ride Hailing Platforms

Large ride-hailing platforms, such as DiDi, Uber and Lyft, connect tens of thousands of vehicles in a city to millions of ride demands throughout the day, providing great promises for improving transportation efficiency through the tasks of order dispatching and vehicle repositioning. Existing studies, however, usually consider the two tasks in simplified settings that hardly address the complex interactions between the two, the real-time fluctuations between supply and demand, and the necessary coordinations due to the large-scale nature of the problem. In this paper we propose a unified value-based dynamic learning framework (V1D3) for tackling both tasks. At the center of the framework is a globally shared value function that is updated continuously using online experiences generated from real-time platform transactions. To improve the sample-efficiency and the robustness, we further propose a novel periodic ensemble method combining the fast online learning with a large-scale offline training scheme that leverages the abundant historical driver trajectory data. This allows the proposed framework to adapt quickly to the highly dynamic environment, to generalize robustly to recurrent patterns and to drive implicit coordinations among the population of managed vehicles. Extensive experiments based on real-world datasets show considerably improvements over other recently proposed methods on both tasks. Particularly, V1D3 outperforms the first prize winners of both dispatching and repositioning tracks in the KDD Cup 2020 RL competition, achieving state-of-the-art results on improving both total driver income and user experience related metrics.

cs.LG↗

arXiv2021

Graph-Based Equilibrium Metrics for Dynamic Supply-Demand Systems with Applications to Ride-sourcing Platforms

How to dynamically measure the local-to-global spatio-temporal coherence between demand and supply networks is a fundamental task for ride-sourcing platforms, such as DiDi. Such coherence measurement is critically important for the quantification of the market efficiency and the comparison of different platform policies, such as dispatching. The aim of this paper is to introduce a graph-based equilibrium metric (GEM) to quantify the distance between demand and supply networks based on a weighted graph structure. We formulate GEM as the optimal objective value of an unbalanced transport problem, which can be efficiently solved by optimizing an equivalent linear programming. We examine how the GEM can help solve three operational tasks of ride-sourcing platforms. The first one is that GEM achieves up to 70.6% reduction in root-mean-square error over the second-best distance measurement for the prediction accuracy. The second one is that the use of GEM for designing order dispatching policy increases answer rate and drivers' revenue for more than 1%, representing a huge improvement in number. The third one is that GEM is to serve as an endpoint for comparing different platform policies in AB test.

math.OC↗

arXiv2020

High-Dimensional Spatial Quantile Function-on-Scalar Regression

This paper develops a novel spatial quantile function-on-scalar regression model, which studies the conditional spatial distribution of a high-dimensional functional response given scalar predictors. With the strength of both quantile regression and copula modeling, we are able to explicitly characterize the conditional distribution of the functional or image response on the whole spatial domain. Our method provides a comprehensive understanding of the effect of scalar covariates on functional responses across different quantile levels and also gives a practical way to generate new images for given covariate values. Theoretically, we establish the minimax rates of convergence for estimating coefficient functions under both fixed and random designs. We further develop an efficient primal-dual algorithm to handle high-dimensional image data. Simulations and real data analysis are conducted to examine the finite-sample performance.

stat.ME↗

arXiv2020

Optimal Passenger-Seeking Policies on E-hailing Platforms Using Markov Decision Process and Imitation Learning

Vacant taxi drivers' passenger seeking process in a road network generates additional vehicle miles traveled, adding congestion and pollution into the road network and the environment. This paper aims to employ a Markov Decision Process (MDP) to model idle e-hailing drivers' optimal sequential decisions in passenger-seeking. Transportation network companies (TNC) or e-hailing (e.g., Didi, Uber) drivers exhibit different behaviors from traditional taxi drivers because e-hailing drivers do not need to actually search for passengers. Instead, they reposition themselves so that the matching platform can match a passenger. Accordingly, we incorporate e-hailing drivers' new features into our MDP model. The reward function used in the MDP model is uncovered by leveraging an inverse reinforcement learning technique. We then use 44,160 Didi drivers' 3-day trajectories to train the model. To validate the effectiveness of the model, a Monte Carlo simulation is conducted to simulate the performance of drivers under the guidance of the optimal policy, which is then compared with the performance of drivers following one baseline heuristic, namely, the local hotspot strategy. The results show that our model is able to achieve a 17.5% improvement over the local hotspot strategy in terms of the rate of return. The proposed MDP model captures the supply-demand ratio considering the fact that the number of drivers in this study is sufficiently large and thus the number of unmatched orders is assumed to be negligible. To better incorporate the competition among multiple drivers into the model, we have also devised and calibrated a dynamic adjustment strategy of the order matching probability.

cs.LG↗

arXiv2019

Cross-trait prediction accuracy of high-dimensional ridge-type estimators in genome-wide association studies

Marginal association summary statistics have attracted great attention in statistical genetics, mainly because the primary results of most genome-wide association studies (GWAS) are produced by marginal screening. In this paper, we study the prediction accuracy of marginal estimator in dense (or sparsity free) high-dimensional settings with $(n,p,m) \to \infty$, $m/n \to γ\in (0,\infty)$, and $p/n \to ω\in (0,\infty)$. We consider a general correlation structure among the $p$ features and allow an unknown subset $m$ of them to be signals. As the marginal estimator can be viewed as a ridge estimator with regularization parameter $λ\to \infty$, we further investigate a class of ridge-type estimators in a unifying framework, including the popular best linear unbiased prediction (BLUP) in genetics. We find that the influence of $λ$ on out-of-sample prediction accuracy heavily depends on $ω$. Though selecting an optimal $λ$ can be important when $p$ and $n$ are comparable, it turns out that the out-of-sample $R^2$ of ridge-type estimators becomes near-optimal for any $λ\in (0,\infty)$ as $ω$ increases. For example, when features are independent, the out-of-sample $R^2$ is always bounded by $1/ω$ from above and is largely invariant to $λ$ given large $ω$ (say, $ω>5$). We also find that in-sample $R^2$ has completely different patterns and depends much more on $λ$ than out-of-sample $R^2$. In practice, our analysis delivers useful messages for genome-wide polygenic risk prediction and computation-accuracy trade-off in dense high-dimensions. We numerically illustrate our results in simulation studies and a real data example.

stat.ME↗

arXiv2019

Partial Least Squares for Functional Joint Models

Many biomedical studies have identified important imaging biomarkers that are associated with both repeated clinical measures and a survival outcome. The functional joint model (FJM) framework, proposed in Li and Luo (2017), investigates the association between repeated clinical measures and survival data, while adjusting for both high-dimensional images and low-dimensional covariates based upon the functional principal component analysis (FPCA). In this paper, we propose a novel algorithm for the estimation of FJM based on the functional partial least squares (FPLS). Our numerical studies demonstrate that, compared to FPCA, the proposed FPLS algorithm can yield more accurate and robust estimation and prediction performance in many important scenarios. We apply the proposed FPLS algorithm to the Alzheimer's Disease Neuroimaging Initiative (ADNI) study. Data used in the preparation of this article were obtained from the ADNI database.

stat.ME↗

arXiv2019

ACE of Space: Estimating Genetic Components of High-Dimensional Imaging Data

It is of great interest to quantify the contributions of genetic variation to brain structure and function, which are usually measured by high-dimensional imaging data (e.g., magnetic resonance imaging). In addition to the variance, the covariance patterns in the genetic effects of a functional phenotype are of biological importance, and covariance patterns have been linked to psychiatric disorders. The aim of this paper is to develop a scalable method to estimate heritability and the non-stationary covariance components in high-dimensional imaging data from twin studies. Our motivating example is from the Human Connectome Project (HCP). Several major big-data challenges arise from estimating the genetic and environmental covariance functions of functional phenotypes extracted from imaging data, such as cortical thickness with 60,000 vertices. Notably, truncating to positive eigenvalues and their eigenfunctions from unconstrained estimators can result in large bias. This motivated our development of a novel estimator ensuring positive semidefiniteness. Simulation studies demonstrate large improvements over existing approaches, both with respect to heritability estimates and covariance estimation. We applied the proposed method to cortical thickness data from the HCP. Our analysis suggests fine-scale differences in covariance patterns, identifying locations in which genetic control is correlated with large areas of the brain and locations where it is highly localized.

stat.AP↗

arXiv2019

On genetic correlation estimation with summary statistics from genome-wide association studies

Genome-wide association studies (GWAS) have been widely used to examine the association between single nucleotide polymorphisms (SNPs) and complex traits, where both the sample size n and the number of SNPs p can be very large. Recently, cross-trait polygenic risk score (PRS) method has gained extremely popular for assessing genetic correlation of complex traits based on GWAS summary statistics (e.g., SNP effect size). However, empirical evidence has shown a common bias phenomenon that even highly significant cross-trait PRS can only account for a very small amount of genetic variance (R^2 often <1%). The aim of this paper is to develop a novel and powerful method to address the bias phenomenon of cross-trait PRS. We theoretically show that the estimated genetic correlation is asymptotically biased towards zero when complex traits are highly polygenic/omnigenic. When all p SNPs are used to construct PRS, we show that the asymptotic bias of PRS estimator is independent of the unknown number of causal SNPs m. We propose a consistent PRS estimator to correct such asymptotic bias. We also develop a novel estimator of genetic correlation which is solely based on two sets of GWAS summary statistics. In addition, we investigate whether or not SNP screening by GWAS p-values can lead to improved estimation and show the effect of overlapping samples among GWAS. Our results may help demystify and tackle the puzzling "missing genetic overlap" phenomenon of cross-trait PRS for dissecting the genetic similarity of closely related heritable traits. We illustrate the finite sample performance of our bias-corrected PRS estimator by using both numerical experiments and the UK Biobank data, in which we assess the genetic correlation between brain white matter tracts and neuropsychiatric disorders.

stat.ME↗

arXiv2019

Sensitivity Analysis of Deep Neural Networks

Deep neural networks (DNNs) have achieved superior performance in various prediction tasks, but can be very vulnerable to adversarial examples or perturbations. Therefore, it is crucial to measure the sensitivity of DNNs to various forms of perturbations in real applications. We introduce a novel perturbation manifold and its associated influence measure to quantify the effects of various perturbations on DNN classifiers. Such perturbations include various external and internal perturbations to input samples and network parameters. The proposed measure is motivated by information geometry and provides desirable invariance properties. We demonstrate that our influence measure is useful for four model building tasks: detecting potential 'outliers', analyzing the sensitivity of model architectures, comparing network sensitivity between training and test sets, and locating vulnerable areas. Experiments show reasonably good performance of the proposed measure for the popular DNN models ResNet50 and DenseNet121 on CIFAR10 and MNIST datasets.

stat.ML↗

arXiv2018

Tensor network factorizations: Relationships between brain structural connectomes and traits

Advanced brain imaging techniques make it possible to measure individuals' structural connectomes in large cohort studies non-invasively. The structural connectome is initially shaped by genetics and subsequently refined by the environment. It is extremely interesting to study relationships between structural connectomes and environment factors or human traits, such as substance use and cognition. Due to limitations in structural connectome recovery, previous studies largely focus on functional connectomes. Questions remain about how well structural connectomes can explain variance in different human traits. Using a state-of-the-art structural connectome processing pipeline and a novel dimensionality reduction technique applied to data from the Human Connectome Project (HCP), we show strong relationships between structural connectomes and various human traits. Our dimensionality reduction approach uses a tensor characterization of the connectome and relies on a generalization of principal components analysis. We analyze over 1100 scans for 1076 subjects from the HCP and the Sherbrooke test-retest data set, as well as $175$ human traits that measure domains including cognition, substance use, motor, sensory and emotion. We find that structural connectomes are associated with many traits. Specifically, fluid intelligence, language comprehension, and motor skills are associated with increased cortical-cortical brain structural connectivity, while the use of alcohol, tobacco, and marijuana are associated with decreased cortical-cortical connectivity.

stat.AP↗

arXiv2018

A Review of Statistical Methods in Imaging Genetics

With the rapid growth of modern technology, many large-scale biomedical studies have been/are being/will be conducted to collect massive datasets with large volumes of multi-modality imaging, genetic, neurocognitive, and clinical information from increasingly large cohorts. Simultaneously extracting and integrating rich and diverse heterogeneous information in neuroimaging and/or genomics from these big datasets could transform our understanding of how genetic variants impact brain structure and function, cognitive function, and brain-related disease risk across the lifespan. Such understanding is critical for diagnosis, prevention, and treatment of numerous complex brain-related disorders (e.g., schizophrenia and Alzheimer). However, the development of analytical methods for the joint analysis of both high-dimensional imaging phenotypes and high-dimensional genetic data, called big data squared (BD$^2$), presents major computational and theoretical challenges for existing analytical methods. Besides the high-dimensional nature of BD$^2$, various neuroimaging measures often exhibit strong spatial smoothness and dependence and genetic markers may have a natural dependence structure arising from linkage disequilibrium. We review some recent developments of various statistical techniques for the joint analysis of BD$^2$, including massive univariate and voxel-wise approaches, reduced rank regression, mixture models, and group sparse multi-task regression. By doing so, we hope that this review may encourage others in the statistical community to enter into this new and exciting field of research.

stat.ME↗

arXiv2017

TPRM: Tensor partition regression models with applications in imaging biomarker detection

Medical imaging studies have collected high dimensional imaging data to identify imaging biomarkers for diagnosis, screening, and prognosis, among many others. These imaging data are often represented in the form of a multi-dimensional array, called a tensor. The aim of this paper is to develop a tensor partition regression modeling (TPRM) framework to establish a relationship between low-dimensional clinical outcomes (e.g., diagnosis) and high dimensional tensor covariates. Our TPRM is a hierarchical model and efficiently integrates four components: (i) a partition model, (ii) a canonical polyadic decomposition model, (iii) a principal components model, and (iv) a generalized linear model with a sparse inducing normal mixture prior. This framework not only reduces ultra-high dimensionality to a manageable level, resulting in efficient estimation, but also optimizes prediction accuracy in the search for informative sub-tensors. Posterior computation proceeds via an efficient Markov chain Monte Carlo algorithm. Simulation shows that TPRM outperforms several other competing methods. We apply TPRM to predict disease status (Alzheimer versus control) by using structural magnetic resonance imaging data obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study.

stat.AP↗

arXiv2017

CLT for linear spectral statistics of large dimensional sample covariance matrices with dependent data

This paper investigates the central limit theorem for linear spectral statistics of high dimensional sample covariance matrices of the form $\mathbf{B}_n=n^{-1}\sum_{j=1}^{n}\mathbf{Q}\mathbf{x}_j\mathbf{x}_j^{*}\mathbf{Q}^{*}$ where $\mathbf{Q}$ is a nonrandom matrix of dimension $p\times k$, and $\{\mathbf{x}_j\}$ is a sequence of independent $k$-dimensional random vector with independent entries, under the assumption that $p/n\to y>0$. A key novelty here is that the dimension $k\ge p$ can be arbitrary, possibly infinity. This new model of sample covariance matrices $\mathbf{B}_n$ covers most of the known models as its special cases. For example, standard sample covariance matrices are obtained with $k=p$ and $\mathbf{Q}=\mathbf{T}_n^{1/2}$ for some positive definite Hermitian matrix $\mathbf{T}_n$. Also with $k=\infty$ our model covers the case of repeated linear processes considered in recent high-dimensional time series literature. The CLT found in this paper substantially generalizes the seminal CLT in Bai and Silverstein (2004). Applications of this new CLT are proposed for testing the structure of a high-dimensional covariance matrix. The derived tests are then used to analyse a large fMRI data set regarding its temporary correlation structure.

math.PR↗

arXiv2017

Bayesian analysis of ambulatory blood pressure dynamics with application to irregularly spaced sparse data

Ambulatory cardiovascular (CV) measurements provide valuable insights into individuals' health conditions in "real-life," everyday settings. Current methods of modeling ambulatory CV data do not consider the dynamic characteristics of the full data set and their relationships with covariates such as caffeine use and stress. We propose a stochastic differential equation (SDE) in the form of a dual nonlinear Ornstein--Uhlenbeck (OU) model with person-specific covariates to capture the morning surge and nighttime dipping dynamics of ambulatory CV data. To circumvent the data analytic constraint that empirical measurements are typically collected at irregular and much larger time intervals than those evaluated in simulation studies of SDEs, we adopt a Bayesian approach with a regularized Brownian Bridge sampler (RBBS) and an efficient multiresolution (MR) algorithm to fit the proposed SDE. The MR algorithm can produce more efficient MCMC samples that is crucial for valid parameter estimation and inference. Using this model and algorithm to data from the Duke Behavioral Investigation of Hypertension Study, results indicate that age, caffeine intake, gender and race have effects on distinct dynamic characteristics of the participants' CV trajectories.

stat.AP↗

arXiv2016

Bayesian spatial transformation models with applications in neuroimaging data

The aim of this paper is to develop a class of spatial transformation models (STM) to spatially model the varying association between imaging measures in a three-dimensional (3D) volume (or 2D surface) and a set of covariates. Our STMs include a varying Box-Cox transformation model for dealing with the issue of non-Gaussian distributed imaging data and a Gaussian Markov Random Field model for incorporating spatial smoothness of the imaging data. Posterior computation proceeds via an efficient Markov chain Monte Carlo algorithm. Simulations and real data analysis demonstrate that the STM significantly outperforms the voxel-wise linear model with Gaussian noise in recovering meaningful geometric patterns. Our STM is able to reveal important brain regions with morphological changes in children with attention deficit hyperactivity disorder.

stat.AP↗