arXiv subjects

Yili Hong

Publications and source records attributed to Yili Hong.

At least 37 records · Page 2Linked to original sources

arXiv2024

A Detailed Historical and Statistical Analysis of the Influence of Hardware Artifacts on SPEC Integer Benchmark Performance

The Standard Performance Evaluation Corporation (SPEC) CPU benchmark has been widely used as a measure of computing performance for decades. The SPEC is an industry-standardized, CPU-intensive benchmark suite and the collective data provide a proxy for the history of worldwide CPU and system performance. Past efforts have not provided or enabled answers to questions such as, how has the SPEC benchmark suite evolved empirically over time and what micro-architecture artifacts have had the most influence on performance? -- have any micro-benchmarks within the suite had undue influence on the results and comparisons among the codes? -- can the answers to these questions provide insights to the future of computer system performance? To answer these questions, we detail our historical and statistical analysis of specific hardware artifacts (clock frequencies, core counts, etc.) on the performance of the SPEC benchmarks since 1995. We discuss in detail several methods to normalize across benchmark evolutions. We perform both isolated and collective sensitivity analyses for various hardware artifacts and we identify one benchmark (libquantum) that had somewhat undue influence on performance outcomes. We also present the use of SPEC data to predict future performance.

cs.CY↗

arXiv2024

Multivariate Functional Clustering with Variable Selection and Application to Sensor Data from Engineering Systems

Multi-sensor data that track system operating behaviors are widely available nowadays from various engineering systems. Measurements from each sensor over time form a curve and can be viewed as functional data. Clustering of these multivariate functional curves is important for studying the operating patterns of systems. One complication in such applications is the possible presence of sensors whose data do not contain relevant information. Hence it is desirable for the clustering method to equip with an automatic sensor selection procedure. Motivated by a real engineering application, we propose a functional data clustering method that simultaneously removes noninformative sensors and groups functional curves into clusters using informative sensors. Functional principal component analysis is used to transform multivariate functional data into a coefficient matrix for data reduction. We then model the transformed data by a Gaussian mixture distribution to perform model-based clustering with variable selection. Three types of penalties, the individual, variable, and group penalties, are considered to achieve automatic variable selection. Extensive simulations are conducted to assess the clustering and variable selection performance of the proposed methods. The application of the proposed methods to an engineering system with multiple sensors shows the promise of the methods and reveals interesting patterns in the sensor data.

stat.ME↗

arXiv2023

Planning Reliability Assurance Tests for Autonomous Vehicles

Artificial intelligence (AI) technology has become increasingly prevalent and transforms our everyday life. One important application of AI technology is the development of autonomous vehicles (AV). However, the reliability of an AV needs to be carefully demonstrated via an assurance test so that the product can be used with confidence in the field. To plan for an assurance test, one needs to determine how many AVs need to be tested for how many miles and the standard for passing the test. Existing research has made great efforts in developing reliability demonstration tests in the other fields of applications for product development and assessment. However, statistical methods have not been utilized in AV test planning. This paper aims to fill in this gap by developing statistical methods for planning AV reliability assurance tests based on recurrent events data. We explore the relationship between multiple criteria of interest in the context of planning AV reliability assurance tests. Specifically, we develop two test planning strategies based on homogeneous and non-homogeneous Poisson processes while balancing multiple objectives with the Pareto front approach. We also offer recommendations for practical use. The disengagement events data from the California Department of Motor Vehicles AV testing program is used to illustrate the proposed assurance test planning methods.

stat.AP↗

arXiv2023

Deep Neural Network Identification of Limnonectes Species and New Class Detection Using Image Data

As is true of many complex tasks, the work of discovering, describing, and understanding the diversity of life on Earth (viz., biological systematics and taxonomy) requires many tools. Some of this work can be accomplished as it has been done in the past, but some aspects present us with challenges which traditional knowledge and tools cannot adequately resolve. One such challenge is presented by species complexes in which the morphological similarities among the group members make it difficult to reliably identify known species and detect new ones. We address this challenge by developing new tools using the principles of machine learning to resolve two specific questions related to species complexes. The first question is formulated as a classification problem in statistics and machine learning and the second question is an out-of-distribution (OOD) detection problem. We apply these tools to a species complex comprising Southeast Asian stream frogs (Limnonectes kuhlii complex) and employ a morphological character (hind limb skin texture) traditionally treated qualitatively in a quantitative and objective manner. We demonstrate that deep neural networks can successfully automate the classification of an image into a known species group for which it has been trained. We further demonstrate that the algorithm can successfully classify an image into a new class if the image does not belong to the existing classes. Additionally, we use the larger MNIST dataset to test the performance of our OOD detection algorithm. We finish our paper with some concluding remarks regarding the application of these methods to species complexes and our efforts to document true biodiversity. This paper has online supplementary materials.

stat.ML↗

arXiv2023

A Spatially Correlated Competing Risks Time-to-Event Model for Supercomputer GPU Failure Data

Graphics processing units (GPUs) are widely used in many high-performance computing (HPC) applications such as imaging/video processing and training deep-learning models in artificial intelligence. GPUs installed in HPC systems are often heavily used, and GPU failures occur during HPC system operations. Thus, the reliability of GPUs is of interest for the overall reliability of HPC systems. The Cray XK7 Titan supercomputer was one of the top ten supercomputers in the world. The failure event times of more than 30,000 GPUs in Titan were recorded and previous data analysis suggested that the failure time of a GPU may be affected by the GPU's connectivity location inside the supercomputer among other factors. In this paper, we conduct in-depth statistical modeling of GPU failure times to study the effect of location on GPU failures under competing risks with covariates and spatially correlated random effects. In particular, two major failure types of GPUs in Titan are considered. The connectivity locations of cabinets are modeled as spatially correlated random effects, and the positions of GPUs inside each cabinet are treated as covariates. A Bayesian framework is used for statistical inference. We also compare different methods of estimation such as the maximum likelihood, which is implemented via an expectation-maximization algorithm. Our results provide interesting insights into GPU failures in HPC systems.

stat.AP↗

arXiv2022

Prediction for Distributional Outcomes in High-Performance Computing I/O Variability

Although high-performance computing (HPC) systems have been scaled to meet the exponentially-growing demand for scientific computing, HPC performance variability remains a major challenge and has become a critical research topic in computer science. Statistically, performance variability can be characterized by a distribution. Predicting performance variability is a critical step in HPC performance variability management and is nontrivial because one needs to predict a distribution function based on system factors. In this paper, we propose a new framework to predict performance distributions. The proposed model is a modified Gaussian process that can predict the distribution function of the input/output (I/O) throughput under a specific HPC system configuration. We also impose a monotonic constraint so that the predicted function is nondecreasing, which is a property of the cumulative distribution function. Additionally, the proposed model can incorporate both quantitative and qualitative input variables. We evaluate the performance of the proposed method by using the IOzone variability data based on various prediction tasks. Results show that the proposed method can generate accurate predictions, and outperform existing methods. We also show how the predicted functional output can be used to generate predictions for a scalar summary of the performance distribution, such as the mean, standard deviation, and quantiles. Our methods can be further used as a surrogate model for HPC system variability monitoring and optimization.

stat.AP↗

arXiv2022

A Little Too Personal: Effects of Standardization versus Personalization on Job Acquisition, Work Completion, and Revenue for Online Freelancers

As more individuals consider permanently working from home, the online labor market continues to grow as an alternative working environment. While the flexibility and autonomy of these online gigs attracts many workers, success depends critically upon self-management and workers' efficient allocation of scarce resources. To achieve this, freelancers may develop alternative work strategies, employing highly standardized schedules and communication patterns while taking on large work volumes, or engaging in smaller numbers of jobs whilst tailoring their activities to build relationships with individual employers. In this study, we consider this contrast in relation to worker communication patterns. We demonstrate the heterogeneous effects of standardization versus personalization across different stages of a project and examine the relative impact on job acquisition, project completion, and earnings. Our findings can inform the design of platforms and various worker support tools for the gig economy.

cs.HC↗

arXiv2022

Design Strategies and Approximation Methods for High-Performance Computing Variability Management

Performance variability management is an active research area in high-performance computing (HPC). We focus on input/output (I/O) variability. To study the performance variability, computer scientists often use grid-based designs (GBDs) to collect I/O variability data, and use mathematical approximation methods to build a prediction model. Mathematical approximation models could be biased particularly if extrapolations are needed. Space-filling designs (SFDs) and surrogate models such as Gaussian process (GP) are popular for data collection and building predictive models. The applicability of SFDs and surrogates in the HPC variability needs investigation. We investigate their applicability in the HPC setting in terms of design efficiency, prediction accuracy, and scalability. We first customize the existing SFDs so that they can be applied in the HPC setting. We conduct a comprehensive investigation of design strategies and the prediction ability of approximation methods. We use both synthetic data simulated from three test functions and the real data from the HPC setting. We then compare different methods in terms of design efficiency, prediction accuracy, and scalability. In synthetic and real data analysis, GP with SFDs outperforms in most scenarios. With respect to approximation models, GP is recommended if the data are collected by SFDs. If data are collected using GBDs, both GP and Delaunay can be considered. With the best choice of approximation method, the performance of SFDs and GBD depends on the property of the underlying surface. For the cases in which SFDs perform better, the number of design points needed for SFDs is about half of or less than that of the GBD to achieve the same prediction accuracy. SFDs that can be tailored to high dimension and non-smooth surface are recommended especially when large numbers of input factors need to be considered in the model.

stat.AP↗

arXiv2022

The Poisson Multinomial Distribution and Its Applications in Voting Theory, Ecological Inference, and Machine Learning

The Poisson multinomial distribution (PMD) describes the distribution of the sum of $n$ independent but non-identically distributed random vectors, in which each random vector is of length $m$ with 0/1 valued elements and only one of its elements can take value 1 with a certain probability. Those probabilities are different for the $m$ elements across the $n$ random vectors, and form an $n \times m$ matrix with row sum equals to 1. We call this $n\times m$ matrix the success probability matrix (SPM). Each SPM uniquely defines a PMD. The PMD is useful in many areas such as, voting theory, ecological inference, and machine learning. The distribution functions of PMD, however, are usually difficult to compute. In this paper, we develop efficient methods to compute the probability mass function (pmf) for the PMD using multivariate Fourier transform, normal approximation, and simulations. We study the accuracy and efficiency of those methods and give recommendations for which methods to use under various scenarios. We also illustrate the use of the PMD via three applications, namely, in voting probability calculation, aggregated data inference, and uncertainty quantification in classification. We build an R package that implements the proposed methods, and illustrate the package with examples.

stat.CO↗

arXiv2021

Statistical Perspectives on Reliability of Artificial Intelligence Systems

Artificial intelligence (AI) systems have become increasingly popular in many areas. Nevertheless, AI technologies are still in their developing stages, and many issues need to be addressed. Among those, the reliability of AI systems needs to be demonstrated so that the AI systems can be used with confidence by the general public. In this paper, we provide statistical perspectives on the reliability of AI systems. Different from other considerations, the reliability of AI systems focuses on the time dimension. That is, the system can perform its designed functionality for the intended period. We introduce a so-called SMART statistical framework for AI reliability research, which includes five components: Structure of the system, Metrics of reliability, Analysis of failure causes, Reliability assessment, and Test planning. We review traditional methods in reliability data analysis and software reliability, and discuss how those existing methods can be transformed for reliability modeling and assessment of AI systems. We also describe recent developments in modeling and analysis of AI reliability and outline statistical research challenges in this area, including out-of-distribution detection, the effect of the training set, adversarial attacks, model accuracy, and uncertainty quantification, and discuss how those topics can be related to AI reliability, with illustrative examples. Finally, we discuss data collection and test planning for AI reliability assessment and how to improve system designs for higher AI reliability. The paper closes with some concluding remarks.

cs.SE↗

arXiv2021

Building Degradation Index with Variable Selection for Multivariate Sensory Data

The modeling and analysis of degradation data have been an active research area in reliability and system health management. As the senor technology advances, multivariate sensory data are commonly collected for the underlying degradation process. However, most existing research on degradation modeling requires a univariate degradation index to be provided. Thus, constructing a degradation index for multivariate sensory data is a fundamental step in degradation modeling. In this paper, we propose a novel degradation index building method for multivariate sensory data. Based on an additive nonlinear model with variable selection, the proposed method can automatically select the most informative sensor signals to be used in the degradation index. The penalized likelihood method with adaptive group penalty is developed for parameter estimation. We demonstrate that the proposed method outperforms existing methods via both simulation studies and analyses of the NASA jet engine sensor data.

stat.AP↗

arXiv2021

Copula-Frailty Models for Recurrent Event Data Based on Monte Carlo EM Algorithm

Multi-type recurrent events are often encountered in medical applications when two or more different event types could repeatedly occur over an observation period. For example, patients may experience recurrences of multi-type nonmelanoma skin cancers in a clinical trial for skin cancer prevention. The aims in those applications are to characterize features of the marginal processes, evaluate covariate effects, and quantify both the within-subject recurrence dependence and the dependence among different event types. We use copula-frailty models to analyze correlated recurrent events of different types. Parameter estimation and inference are carried out by using a Monte Carlo expectation-maximization (MCEM) algorithm, which can handle a relatively large (i.e., three or more) number of event types. Performances of the proposed methods are evaluated via extensive simulation studies. The developed methods are used to model the recurrences of skin cancer with different types.

stat.ME↗

arXiv2021

Modeling Bivariate Geyser Eruption System with Covariate-Adjusted Recurrent Event Process

Geyser eruption is one of the most popular signature attractions at the Yellowstone National Park. The interdependence of geyser eruptions and impacts of covariates are of interest to researchers in geyser studies. In this paper, we propose a parametric covariate-adjusted recurrent event model for estimating the eruption gap time. We describe a general bivariate recurrent event process, where a bivariate lognormal distribution and a Gumbel copula with different marginal distributions are used to model an interdependent dual-type event system. The maximum likelihood approach is used to estimate model parameters. The proposed method is applied to analyzing the Yellowstone geyser eruption data for a bivariate geyser system and offers a deeper understanding of the event occurrence mechanism of individual events as well as the system as a whole. A comprehensive simulation study is conducted to evaluate the performance of the proposed method.

stat.AP↗

arXiv2021

Reliability Analysis of Artificial Intelligence Systems Using Recurrent Events Data from Autonomous Vehicles

Artificial intelligence (AI) systems have become increasingly common and the trend will continue. Examples of AI systems include autonomous vehicles (AV), computer vision, natural language processing, and AI medical experts. To allow for safe and effective deployment of AI systems, the reliability of such systems needs to be assessed. Traditionally, reliability assessment is based on reliability test data and the subsequent statistical modeling and analysis. The availability of reliability data for AI systems, however, is limited because such data are typically sensitive and proprietary. The California Department of Motor Vehicles (DMV) oversees and regulates an AV testing program, in which many AV manufacturers are conducting AV road tests. Manufacturers participating in the program are required to report recurrent disengagement events to California DMV. This information is being made available to the public. In this paper, we use recurrent disengagement events as a representation of the reliability of the AI system in AV, and propose a statistical framework for modeling and analyzing the recurrent events data from AV driving tests. We use traditional parametric models in software reliability and propose a new nonparametric model based on monotonic splines to describe the event process. We develop inference procedures for selecting the best models, quantifying uncertainty, and testing heterogeneity in the event process. We then analyze the recurrent events data from four AV manufacturers, and make inferences on the reliability of the AI systems in AV. We also describe how the proposed analysis can be applied to assess the reliability of other AI systems.

cs.AI↗

arXiv2021

Sequential Design of Computer Experiments with Quantitative and Qualitative Factors in Applications to HPC Performance Optimization

Computer experiments with both qualitative and quantitative factors are widely used in many applications. Motivated by the emerging need of optimal configuration in the high-performance computing (HPC) system, this work proposes a sequential design, denoted as adaptive composite exploitation and exploration (CEE), for optimization of computer experiments with qualitative and quantitative factors. The proposed adaptive CEE method combines the predictive mean and standard deviation based on the additive Gaussian process to achieve a meaningful balance between exploitation and exploration for optimization. Moreover, the adaptiveness of the proposed sequential procedure allows the selection of next design point from the adaptive design region. Theoretical justification of the adaptive design region is provided. The performance of the proposed method is evaluated by several numerical examples in simulations. The case study of HPC performance optimization further elaborates the merits of the proposed method.

cs.DC↗

arXiv2020

Prediction of High-Performance Computing Input/Output Variability and Its Application to Optimization for System Configurations

Performance variability is an important measure for a reliable high performance computing (HPC) system. Performance variability is affected by complicated interactions between numerous factors, such as CPU frequency, the number of input/output (IO) threads, and the IO scheduler. In this paper, we focus on HPC IO variability. The prediction of HPC variability is a challenging problem in the engineering of HPC systems and there is little statistical work on this problem to date. Although there are many methods available in the computer experiment literature, the applicability of existing methods to HPC performance variability needs investigation, especially, when the objective is to predict performance variability both in interpolation and extrapolation settings. A data analytic framework is developed to model data collected from large-scale experiments. Various promising methods are used to build predictive models for the variability of HPC systems. We evaluate the performance of the methods by measuring prediction accuracy at previously unseen system configurations. We also discuss a methodology for optimizing system configurations that uses the estimated variability map. The findings from method comparisons and developed tool sets in this paper yield new insights into existing statistical methods and can be beneficial for the practice of HPC variability management. This paper has supplementary materials online.

cs.DC↗

arXiv2020

The Unintended Consequences of Stay-at-Home Policies on Work Outcomes: The Impacts of Lockdown Orders on Content Creation

The COVID-19 pandemic has posed an unprecedented challenge to individuals around the globe. To mitigate the spread of the virus, many states in the U.S. issued lockdown orders to urge their residents to stay at their homes, avoid get-togethers, and minimize physical interactions. While many offline workers are experiencing significant challenges performing their duties, digital technologies have provided ample tools for individuals to continue working and to maintain their productivity. Although using digital platforms to build resilience in remote work is effective, other aspects of remote work (beyond the continuation of work) should also be considered in gauging true resilience. In this study, we focus on content creators, and investigate how restrictions in individual's physical environment impact their online content creation behavior. Exploiting a natural experimental setting wherein four states issued state-wide lockdown orders on the same day whereas five states never issued a lockdown order, and using a unique dataset collected from a short video-sharing social media platform, we study the impact of lockdown orders on content creators' behaviors in terms of content volume, content novelty, and content optimism. We combined econometric methods (difference-in-differences estimations of a matched sample) with machine learning-based natural language processing to show that on average, compared to the users residing in non-lockdown states, the users residing in lockdown states create more content after the lockdown order enforcement. However, we find a decrease in the novelty level and optimism of the content generated by the latter group. Our findings have important contributions to the digital resilience literature and shed light on managers' decision-making process related to the adjustment of employees' work mode in the long run.

econ.GN↗

arXiv2020

Investigating the Robustness of Artificial Intelligent Algorithms with Mixture Experiments

Artificial intelligent (AI) algorithms, such as deep learning and XGboost, are used in numerous applications including computer vision, autonomous driving, and medical diagnostics. The robustness of these AI algorithms is of great interest as inaccurate prediction could result in safety concerns and limit the adoption of AI systems. In this paper, we propose a framework based on design of experiments to systematically investigate the robustness of AI classification algorithms. A robust classification algorithm is expected to have high accuracy and low variability under different application scenarios. The robustness can be affected by a wide range of factors such as the imbalance of class labels in the training dataset, the chosen prediction algorithm, the chosen dataset of the application, and a change of distribution in the training and test datasets. To investigate the robustness of AI classification algorithms, we conduct a comprehensive set of mixture experiments to collect prediction performance results. Then statistical analyses are conducted to understand how various factors affect the robustness of AI classification algorithms. We summarize our findings and provide suggestions to practitioners in AI applications.

stat.ML↗