arXiv subjects

Xiao Fang

Publications and source records attributed to Xiao Fang.

At least 55 records · Page 3Linked to original sources

arXiv2024

Bias of AI-Generated Content: An Examination of News Produced by Large Language Models

Large language models (LLMs) have the potential to transform our lives and work through the content they generate, known as AI-Generated Content (AIGC). To harness this transformation, we need to understand the limitations of LLMs. Here, we investigate the bias of AIGC produced by seven representative LLMs, including ChatGPT and LLaMA. We collect news articles from The New York Times and Reuters, both known for their dedication to provide unbiased news. We then apply each examined LLM to generate news content with headlines of these news articles as prompts, and evaluate the gender and racial biases of the AIGC produced by the LLM by comparing the AIGC and the original news articles. We further analyze the gender bias of each LLM under biased prompts by adding gender-biased messages to prompts constructed from these news headlines. Our study reveals that the AIGC produced by each examined LLM demonstrates substantial gender and racial biases. Moreover, the AIGC generated by each LLM exhibits notable discrimination against females and individuals of the Black race. Among the LLMs, the AIGC generated by ChatGPT demonstrates the lowest level of bias, and ChatGPT is the sole model capable of declining content generation when provided with biased prompts.

cs.AI↗

arXiv2024

Machine Learning LSST 3x2pt analyses -- forecasting the impact of systematics on cosmological constraints using neural networks

Validating modeling choices through simulated analyses and quantifying the impact of different systematic effects will form a major computational bottleneck in the preparation for 3$\times$2 analysis with Stage-IV surveys such as Vera Rubin Observatory's Legacy Survey of Space and Time (LSST). We can significantly reduce the computational requirements by using machine learning based emulators, which allow us to run fast inference while maintaining the full realism of the data analysis pipeline. In this paper, we use such an emulator to run simulated 3$\times$2 (cosmic shear, galaxy-galaxy lensing, and galaxy clustering) analyses for mock LSST-Y1/Y3/Y6/Y10 surveys and study the impact of various systematic effects (galaxy bias, intrinsic alignment, baryonic physics, shear calibration and photo-$z$ uncertainties). Closely following the DESC Science Requirement Document (with several updates) our main findings are: {\it a)} The largest contribution to the `systematic error budget' of LSST 3$\times$2 analysis comes from galaxy bias uncertainties, while the contribution of baryonic and shear calibration uncertainties are significantly less important. {\it b)} Tighter constraints on intrinsic alignment and photo-$z$ parameters can improve cosmological constraints noticeably, which illustrates synergies of LSST and spectroscopic surveys. {\it c)} The scale cuts adopted in the DESC SRD may be too conservative and pushing to smaller scales can increase cosmological information significantly. {\it d)} We investigate the impact of photo-$z$ outliers on 3$\times$2 pt analysis and find that we need to determine the outlier fraction to within $5-10\%$ accuracy to ensure robust cosmological analysis. We caution that these findings depend on analysis choices (parameterizations, priors, scale cuts) and can change for different settings.

astro-ph.CO↗

arXiv2024

A Novel Shortest Path Query Algorithm Based on Optimized Adaptive Topology Structure

Urban rail transit is a fundamental component of public transportation, however, commonly station-based path search algorithms often overlook the impact of transfer times on search results, leading to decreased accuracy. To solve this problem, this paper proposes a novel shortest path query algorithm based on adaptive topology optimization called the Adaptive Topology Extension Road Network Structure (ATEN). This algorithm categorizes transfer stations into different types and treats travel time and transfer time equivalently as weights for edges in the topological graph. The proposed algorithm introduces virtual stations to differentiate between pedestrian paths and train paths, eliminating the need for additional operations on transfer stations. The algorithm controls the extent of expansion in the urban rail transit topology, overcoming query errors caused by mishandling of transfer stations in the existing algorithm. Finally, a series of simulation experiments were conducted on Beijing's urban rail transit network to validate both correctness and efficiency of the proposed adaptive topology optimization algorithm. The results demonstrate significant advantages compared to existing similar algorithms.

cs.CE↗

arXiv2024

Bernstein von-Mises Theorem for g-prior and nonlocal prior

The paper develops Bernstein von Mises Theorem under hierarchical $g$ -priors for linear regression models. The results are obtained both when the error variance is known, and also when it is unknown. An inverse gamma prior is attached to the error variance in the later case. The paper also demonstrates some connection between the total variation and $α$-divergence measures.

math.ST↗

arXiv2024

Data Valuation for Vertical Federated Learning: A Model-free and Privacy-preserving Method

Vertical Federated learning (VFL) is a promising paradigm for predictive analytics, empowering an organization (i.e., task party) to enhance its predictive models through collaborations with multiple data suppliers (i.e., data parties) in a decentralized and privacy-preserving way. Despite the fast-growing interest in VFL, the lack of effective and secure tools for assessing the value of data owned by data parties hinders the application of VFL in business contexts. In response, we propose FedValue, a privacy-preserving, task-specific but model-free data valuation method for VFL, which consists of a data valuation metric and a federated computation method. Specifically, we first introduce a novel data valuation metric, namely MShapley-CMI. The metric evaluates a data party's contribution to a predictive analytics task without the need of executing a machine learning model, making it well-suited for real-world applications of VFL. Next, we develop an innovative federated computation method that calculates the MShapley-CMI value for each data party in a privacy-preserving manner. Extensive experiments conducted on six public datasets validate the efficacy of FedValue for data valuation in the context of VFL. In addition, we illustrate the practical utility of FedValue with a case study involving federated movie recommendations.

cs.LG↗

arXiv2024

Second-order Approximation of Exponential Random Graph Models

Exponential random graph models (ERGMs) are flexible probability models allowing edge dependency. However, it is known that, to a first-order approximation, many ERGMs behave like Erdös-Rényi random graphs, where edges are independent. In this paper, to distinguish ERGMs from Erdös-Rényi random graphs, we consider second-order approximations of ERGMs using two-stars and triangles. We prove that the second-order approximation indeed achieves second-order accuracy in the triangle-free case. The new approximation is formally obtained by Hoeffding decomposition and rigorously justified using Stein's method.

math.PR↗

arXiv2023

A Non-Uniform Low-Light Image Enhancement Method with Multi-Scale Attention Transformer and Luminance Consistency Loss

Low-light image enhancement aims to improve the perception of images collected in dim environments and provide high-quality data support for image recognition tasks. When dealing with photos captured under non-uniform illumination, existing methods cannot adaptively extract the differentiated luminance information, which will easily cause over-exposure and under-exposure. From the perspective of unsupervised learning, we propose a multi-scale attention Transformer named MSATr, which sufficiently extracts local and global features for light balance to improve the visual quality. Specifically, we present a multi-scale window division scheme, which uses exponential sequences to adjust the window size of each layer. Within different-sized windows, the self-attention computation can be refined, ensuring the pixel-level feature processing capability of the model. For feature interaction across windows, a global transformer branch is constructed to provide comprehensive brightness perception and alleviate exposure problems. Furthermore, we propose a loop training strategy, using the diverse images generated by weighted mixing and a luminance consistency loss to improve the model's generalization ability effectively. Extensive experiments on several benchmark datasets quantitatively and qualitatively prove that our MSATr is superior to state-of-the-art low-light image enhancement methods, and the enhanced images have more natural brightness and outstanding details. The code is released at https://github.com/fang001021/MSATr.

cs.CV↗

arXiv2023

Cosmology from weak lensing, galaxy clustering, CMB lensing and tSZ: I. 10x2pt Modelling Methodology

The overlap of galaxy surveys and CMB experiments presents an ideal opportunity for joint cosmological dataset analyses. In this paper we develop a halo-model-based method for the first joint analysis combining these two experiments using 10 correlated two-point functions (10x2pt) derived from galaxy position, galaxy shear, CMB lensing convergence, and Compton-y fields. We explore this method using the Vera Rubin Observatory Legacy Survey of Space and Time (LSST) and the Simons Observatory (SO) as examples. We find such LSSxCMB joint analyses lead to significant improvement in Figure-of-Merit of $Ω_m$ and $S_8$ over the constraints from using LSS-only probes within $Λ$CDM. We identify that the shear-$y$ and $y$-$y$ correlations are the most valuable additions when tSZ is included. We further identify the dominant sources of halo model uncertainties in the small-scale modelling, and investigate the impact of halo self-calibration due to the inclusion of small-scale tSZ information.

astro-ph.CO↗

arXiv2023

Constraining Baryonic Physics with DES Y1 and Planck data -- Combining Galaxy Clustering, Weak Lensing, and CMB Lensing

We constrain cosmology and baryonic feedback scenarios with a joint analysis of weak lensing, galaxy clustering, cosmic microwave background (CMB) lensing, and their cross-correlations (so-called 6$\times$2) using data from the Dark Energy Survey (DES) Y1 and the Planck satellite mission. Noteworthy features of our 6$\times$2 pipeline are: We extend CMB lensing cross-correlation measurements to a band surrounding the DES Y1 footprint (a $\sim 25\%$ gain in pairs), and we develop analytic covariance capabilities that account for different footprints and all cross-terms in the 6$\times$2 analysis. We also measure the DES Y1 cosmic shear two-point correlation function (2PCF) down to $0.^\prime 25$, but find that going below $2.^\prime 5$ does not increase cosmological information due to shape noise. We model baryonic physics uncertainties via the amplitude of Principal Components (PCs) derived from a set of hydro-simulations. Given our statistical uncertainties, varying the first PC amplitude $Q_1$ is sufficient to model small-scale cosmic shear 2PCF. For DES Y1+Planck 6$\times$2 we find $S_8=0.799\pm0.016$, comparable to the 5$\times$2 result of DES Y3+SPT/Planck $S_8=0.773\pm0.016$. Combined with our most informative cosmology priors -- baryon acoustic oscillation (BAO), big bang nucleosynthesis (BBN), type Ia supernovae (SNe Ia), and Planck 2018 EE+lowE, we measure $S_8=0.817\pm 0.011$. Regarding baryonic physics constraints, our 6$\times$2 analysis finds $Q_1=2.8\pm1.8$. Combined with the aforementioned priors, it improves the constraint to $Q_1=3.5\pm1.3$. For comparison, the strongest feedback scenario considered in this paper, the cosmo-OWLS AGN ($ΔT_\mathrm{heat}=10^{8.7}$ K), corresponds to $Q_1=5.84$.

astro-ph.CO↗

arXiv2023

Proactive Resource Request for Disaster Response: A Deep Learning-based Optimization Model

Disaster response is critical to save lives and reduce damages in the aftermath of a disaster. Fundamental to disaster response operations is the management of disaster relief resources. To this end, a local agency (e.g., a local emergency resource distribution center) collects demands from local communities affected by a disaster, dispatches available resources to meet the demands, and requests more resources from a central emergency management agency (e.g., Federal Emergency Management Agency in the U.S.). Prior resource management research for disaster response overlooks the problem of deciding optimal quantities of resources requested by a local agency. In response to this research gap, we define a new resource management problem that proactively decides optimal quantities of requested resources by considering both currently unfulfilled demands and future demands. To solve the problem, we take salient characteristics of the problem into consideration and develop a novel deep learning method for future demand prediction. We then formulate the problem as a stochastic optimization model, analyze key properties of the model, and propose an effective solution method to the problem based on the analyzed properties. We demonstrate the superior performance of our method over prevalent existing methods using both real world and simulated data. We also show its superiority over prevalent existing methods in a multi-stakeholder and multi-objective setting through simulations.

cs.LG↗

arXiv2023

Sharp High-dimensional Central Limit Theorems for Log-concave Distributions

Let $X_1,\dots,X_n$ be i.i.d. log-concave random vectors in $\mathbb R^d$ with mean 0 and covariance matrix $Σ$. We study the problem of quantifying the normal approximation error for $W=n^{-1/2}\sum_{i=1}^nX_i$ with explicit dependence on the dimension $d$. Specifically, without any restriction on $Σ$, we show that the approximation error over rectangles in $\mathbb R^d$ is bounded by $C(\log^{13}(dn)/n)^{1/2}$ for some universal constant $C$. Moreover, if the Kannan-Lovász-Simonovits (KLS) spectral gap conjecture is true, this bound can be improved to $C(\log^{3}(dn)/n)^{1/2}$. This improved bound is optimal in terms of both $n$ and $d$ in the regime $\log n=O(\log d)$. We also give $p$-Wasserstein bounds with all $p\geq2$ and a Cramér type moderate deviation result for this normal approximation error, and they are all optimal under the KLS conjecture. To prove these bounds, we develop a new Gaussian coupling inequality that gives almost dimension-free bounds for projected versions of $p$-Wasserstein distance for every $p\geq2$. We prove this coupling inequality by combining Stein's method and Eldan's stochastic localization procedure.

math.PR↗

arXiv2023

High-dimensional Central Limit Theorems by Stein's Method in the Degenerate Case

In the literature of high-dimensional central limit theorems, there is a gap between results for general limiting correlation matrix $Σ$ and the strongly non-degenerate case. For the general case where $Σ$ may be degenerate, under certain light-tail conditions, when approximating a normalized sum of $n$ independent random vectors by the Gaussian distribution $N(0,Σ)$ in multivariate Kolmogorov distance, the best-known error rate has been $O(n^{-1/4})$, subject to logarithmic factors of the dimension. For the strongly non-degenerate case, that is, when the minimum eigenvalue of $Σ$ is bounded away from 0, the error rate can be improved to $O(n^{-1/2})$ up to a $\log n$ factor. In this paper, we show that the $O(n^{-1/2})$ rate up to a $\log n$ factor can still be achieved in the degenerate case, provided that the minimum eigenvalue of the limiting correlation matrix of any three components is bounded away from 0. We prove our main results using Stein's method in conjunction with previously unexplored inequalities for the integral of the first three derivatives of the standard Gaussian density over convex polytopes. These inequalities were previously known only for hyperrectangles. Our proof demonstrates the connection between the three-components condition and the third moment Berry--Esseen bound.

math.PR↗

arXiv2023

Simulation Software of the JUNO Experiment

The Jiangmen Underground Neutrino Observatory (JUNO) is a multi-purpose experiment, under construction in southeast China, that is designed to determine the neutrino mass ordering and precisely measure neutrino oscillation parameters. Monte Carlo simulation plays an important role for JUNO detector design, detector commissioning, offline data processing, and physics processing. The JUNO experiment has the world's largest liquid scintillator detector instrumented with many thousands of PMTs. The broad energy range of interest, long lifetime, and the large scale present data processing challenges across all areas. This paper describes the JUNO simulation software, highlighting the challenges of JUNO simulation and solutions to meet these challenges, including such issues as support for time-correlated analysis, event mixing, event correlation and handling the simulation of many millions of optical photons.

hep-ex↗

arXiv2023

Rethinking Boundary Detection in Deep Learning Models for Medical Image Segmentation

Medical image segmentation is a fundamental task in the community of medical image analysis. In this paper, a novel network architecture, referred to as Convolution, Transformer, and Operator (CTO), is proposed. CTO employs a combination of Convolutional Neural Networks (CNNs), Vision Transformer (ViT), and an explicit boundary detection operator to achieve high recognition accuracy while maintaining an optimal balance between accuracy and efficiency. The proposed CTO follows the standard encoder-decoder segmentation paradigm, where the encoder network incorporates a popular CNN backbone for capturing local semantic information, and a lightweight ViT assistant for integrating long-range dependencies. To enhance the learning capacity on boundary, a boundary-guided decoder network is proposed that uses a boundary mask obtained from a dedicated boundary detection operator as explicit supervision to guide the decoding learning process. The performance of the proposed method is evaluated on six challenging medical image segmentation datasets, demonstrating that CTO achieves state-of-the-art accuracy with a competitive model complexity.

cs.CV↗

arXiv2023

Primordial non-Gaussianity with Angular correlation function: Integral constraint and validation for DES

Local primordial non-Gaussianity (PNG) is a promising observable of the underlying physics of inflation, characterised by $f_{\rm NL}^{\rm loc}$. We present the methodology to measure $f_{\rm NL}^{\rm loc}$ from the Dark Energy Survey (DES) data using the 2-point angular correlation function (ACF) with scale-dependent bias. One of the focuses of the work is the integral constraint. This condition appears when estimating the mean number density of galaxies from the data and is key in obtaining unbiased $f_{\rm NL}^{\rm loc}$ constraints. The methods are analysed for two types of simulations: $\sim 246$ GOLIAT-PNG N-body small area simulations with $f_{\rm NL}$ equal to -100 and 100, and 1952 Gaussian ICE-COLA mocks with $f_{\rm NL}=0$ that follow the DES angular and redshift distribution. We use the ensemble of GOLIAT-PNG mocks to show the importance of the integral constraint when measuring PNG, where we recover the fiducial values of $f_{\rm NL}$ within the $1σ$ when including the integral constraint. In contrast, we found a bias of $Δf_{\rm NL}\sim 100$ when not including it. For a DES-like scenario, we forecast a bias of $Δf_{\rm NL} \sim 23$, equivalent to $1.8σ$, when not using the IC for a fiducial value of $f_{\rm NL}=100$. We use the ICE-COLA mocks to validate our analysis in a realistic DES-like setup finding it robust to different analysis choices: best-fit estimator, the effect of IC, BAO damping, covariance, and scale choices. We forecast a measurement of $f_{\rm NL}$ within $σ(f_{\rm NL})=31$ when using the DES-Y3 BAO sample, with the ACF in the $1\ {\rm deg}<θ<20\ {\rm deg}$ range.

astro-ph.CO↗

arXiv2022

Edgeworth Expansion by Stein's Method

Edgeworth expansion provides higher-order corrections to the normal approximation for a probability distribution. The classical proof of Edgeworth expansion is via characteristic functions. As a powerful method for distributional approximations, Stein's method has also been used to prove Edgeworth expansion results. However, these results assume that either the test function is smooth (which excludes indicator functions of the half line) or that the random variables are continuous (which excludes random variables having only a continuous component). Thus, how to recover the classical Edgeworth expansion result using Stein's method has remained an open problem. In this paper, we develop Stein's method for two-term Edgeworth expansions in a general case. Our approach involves repeated use of Stein equations, Stein identities via Stein kernels, and a replacement argument.

math.PR↗

arXiv2022

Diversity Preference-Aware Link Recommendation for Online Social Networks

Link recommendation, which recommends links to connect unlinked online social network users, is a fundamental social network analytics problem with ample business implications. Existing link recommendation methods tend to recommend similar friends to a user but overlook the user's diversity preference, although social psychology theories suggest the criticality of diversity preference to link recommendation performance. In recommender systems, a field related to link recommendation, a number of diversification methods have been proposed to improve the diversity of recommended items. Nevertheless, diversity preference is distinct from diversity studied by diversification methods. To address these research gaps, we define and operationalize the concept of diversity preference for link recommendation and propose a new link recommendation problem: the diversity preference-aware link recommendation problem. We then analyze key properties of the new link recommendation problem and develop a novel link recommendation method to solve the problem. Using two large-scale online social network data sets, we conduct extensive empirical evaluations to demonstrate the superior performance of our method over representative diversification methods adapted for link recommendation as well as state-of-the-art link recommendation methods.

cs.LG↗

arXiv2022

Exploiting Expert Knowledge for Assigning Firms to Industries: A Novel Deep Learning Method

Industry assignment, which assigns firms to industries according to a predefined Industry Classification System (ICS), is fundamental to a large number of critical business practices, ranging from operations and strategic decision making by firms to economic analyses by government agencies. Three types of expert knowledge are essential to effective industry assignment: definition-based knowledge (i.e., expert definitions of each industry), structure-based knowledge (i.e., structural relationships among industries as specified in an ICS), and assignment-based knowledge (i.e., prior firm-industry assignments performed by domain experts). Existing industry assignment methods utilize only assignment-based knowledge to learn a model that classifies unassigned firms to industries, and overlook definition-based and structure-based knowledge. Moreover, these methods only consider which industry a firm has been assigned to, but ignore the time-specificity of assignment-based knowledge, i.e., when the assignment occurs. To address the limitations of existing methods, we propose a novel deep learning-based method that not only seamlessly integrates the three types of knowledge for industry assignment but also takes the time-specificity of assignment-based knowledge into account. Methodologically, our method features two innovations: dynamic industry representation and hierarchical assignment. The former represents an industry as a sequence of time-specific vectors by integrating the three types of knowledge through our proposed temporal and spatial aggregation mechanisms. The latter takes industry and firm representations as inputs, computes the probability of assigning a firm to different industries, and assigns the firm to the industry with the highest probability.

cs.LG↗