arXiv subjects

Yiming Xu

Publications and source records attributed to Yiming Xu.

At least 109 records · Page 6Linked to original sources

arXiv2022

Probabilistic methods for approximate archetypal analysis

Archetypal analysis is an unsupervised learning method for exploratory data analysis. One major challenge that limits the applicability of archetypal analysis in practice is the inherent computational complexity of the existing algorithms. In this paper, we provide a novel approximation approach to partially address this issue. Utilizing probabilistic ideas from high-dimensional geometry, we introduce two preprocessing techniques to reduce the dimension and representation cardinality of the data, respectively. We prove that provided the data is approximately embedded in a low-dimensional linear subspace and the convex hull of the corresponding representations is well approximated by a polytope with a few vertices, our method can effectively reduce the scaling of archetypal analysis. Moreover, the solution of the reduced problem is near-optimal in terms of prediction errors. Our approach can be combined with other acceleration techniques to further mitigate the intrinsic complexity of archetypal analysis. We demonstrate the usefulness of our results by applying our method to summarize several moderately large-scale datasets.

stat.CO↗

arXiv2022

A General Pairwise Comparison Model for Extremely Sparse Networks

Statistical inference using pairwise comparison data is an effective approach to analyzing large-scale sparse networks. In this paper, we propose a general framework to model the mutual interactions in a network, which enjoys ample flexibility in terms of model parametrization. Under this setup, we show that the maximum likelihood estimator for the latent score vector of the subjects is uniformly consistent under a near-minimal condition on network sparsity. This condition is sharp in terms of the leading order asymptotics describing the sparsity. Our analysis utilizes a novel chaining technique and illustrates an important connection between graph topology and model consistency. Our results guarantee that the maximum likelihood estimator is justified for estimation in large-scale pairwise comparison networks where data are asymptotically deficient. Simulation studies are provided in support of our theoretical findings.

stat.ML↗

arXiv2022

A bandit-learning approach to multifidelity approximation

Multifidelity approximation is an important technique in scientific computation and simulation. In this paper, we introduce a bandit-learning approach for leveraging data of varying fidelities to achieve precise estimates of the parameters of interest. Under a linear model assumption, we formulate a multifidelity approximation as a modified stochastic bandit, and analyze the loss for a class of policies that uniformly explore each model before exploiting. Utilizing the estimated conditional mean-squared error, we propose a consistent algorithm, adaptive Explore-Then-Commit (AETC), and establish a corresponding trajectory-wise optimality result. These results are then extended to the case of vector-valued responses, where we demonstrate that the algorithm is efficient without the need to worry about estimating high-dimensional parameters. The main advantage of our approach is that we require neither hierarchical model structure nor \textit{a priori} knowledge of statistical information (e.g., correlations) about or between models. Instead, the AETC algorithm requires only knowledge of which model is a trusted high-fidelity model, along with (relative) computational cost estimates of querying each model. Numerical experiments are provided at the end to support our theoretical findings.

math.NA↗

arXiv2021

Stable and Compact Face Recognition via Unlabeled Data Driven Sparse Representation-Based Classification

Sparse representation-based classification (SRC) has attracted much attention by casting the recognition problem as simple linear regression problem. SRC methods, however, still is limited to enough labeled samples per category, insufficient use of unlabeled samples, and instability of representation. For tackling these problems, an unlabeled data driven inverse projection pseudo-full-space representation-based classification model is proposed with low-rank sparse constraints. The proposed model aims to mine the hidden semantic information and intrinsic structure information of all available data, which is suitable for few labeled samples and proportion imbalance between labeled samples and unlabeled samples problems in frontal face recognition. The mixed Gauss-Seidel and Jacobian ADMM algorithm is introduced to solve the model. The convergence, representation capability and stability of the model are analyzed. Experiments on three public datasets show that the proposed LR-S-PFSRC model achieves stable results, especially for proportion imbalance of samples.

cs.CV↗

arXiv2021

A Global to Local Double Embedding Method for Multi-person Pose Estimation

Multi-person pose estimation is a fundamental and challenging problem to many computer vision tasks. Most existing methods can be broadly categorized into two classes: top-down and bottom-up methods. Both of the two types of methods involve two stages, namely, person detection and joints detection. Conventionally, the two stages are implemented separately without considering their interactions between them, and this may inevitably cause some issue intrinsically. In this paper, we present a novel method to simplify the pipeline by implementing person detection and joints detection simultaneously. We propose a Double Embedding (DE) method to complete the multi-person pose estimation task in a global-to-local way. DE consists of Global Embedding (GE) and Local Embedding (LE). GE encodes different person instances and processes information covering the whole image and LE encodes the local limbs information. GE functions for the person detection in top-down strategy while LE connects the rest joints sequentially which functions for joint grouping and information processing in A bottom-up strategy. Based on LE, we design the Mutual Refine Machine (MRM) to reduce the prediction difficulty in complex scenarios. MRM can effectively realize the information communicating between keypoints and further improve the accuracy. We achieve the competitive results on benchmarks MSCOCO, MPII and CrowdPose, demonstrating the effectiveness and generalization ability of our method.

cs.CV↗

arXiv2021

Randomized weakly admissible meshes

A weakly admissible mesh (WAM) on a continuum real-valued domain is a sequence of discrete grids such that the discrete maximum norm of polynomials on the grid is comparable to the supremum norm of polynomials on the domain. The asymptotic rate of growth of the grid sizes and of the comparability constants must grow in a controlled manner. In this paper, we generalize the notion of a WAM to a hierarchical subspaces of not necessarily polynomial functions, and we analyze particular strategies for random sampling as a technique for generating WAM's. Our main results show that WAM's and their stronger variant, admissible meshes (AM's), can be generated by random sampling, and our analysis provides concrete estimates for growth of both the meshes and the discrete-continuum comparability constants.

math.NA↗

arXiv2021

Estimating Wildfire Evacuation Decision and Departure Timing Using Large-Scale GPS Data

With increased frequency and intensity due to climate change, wildfires have become a growing global concern. This creates severe challenges for fire and emergency services as well as communities in the wildland-urban interface (WUI). To reduce wildfire risk and enhance the safety of WUI communities, improving our understanding of wildfire evacuation is a pressing need. To this end, this study proposes a new methodology to analyze human behavior during wildfires by leveraging a large-scale GPS dataset. This methodology includes a home-location inference algorithm and an evacuation-behavior inference algorithm, to systematically identify different groups of wildfire evacuees (i.e., self-evacuee, shadow evacuee, evacuee under warning, and ordered evacuee). We applied the methodology to the 2019 Kincade Fire in Sonoma County, CA. We found that among all groups of evacuees, self-evacuees and shadow evacuees accounted for more than half of the evacuees during the Kincade Fire. The results also show that inside of the evacuation warning/order zones, the total evacuation compliance rate was around 46% among all the categorized people. The findings of this study can be used by emergency managers and planners to better target public outreach campaigns, training protocols, and emergency communication strategies to prepare WUI households for future wildfire events.

cs.CY↗

arXiv2021

Do e-scooters fill mobility gaps and promote equity before and during COVID-19? A spatiotemporal analysis using open big data

The growing popularity of e-scooters and their rapid expansion across urban streets has attracted widespread attention. A major policy question is whether e-scooters substitute existing mobility options or fill the service gaps left by them. This study addresses this question by analyzing the spatiotemporal patterns of e-scooter service availability and use in Washington DC, focusing on their spatial relationships with public transit and bikesharing. Results from an analysis of three open big datasets suggest that e-scooters have both competing and complementary effects on transit and bikesharing services. The supply of e-scooters significantly overlaps with the service areas of transit and bikesharing, and we classify a majority of e-scooter trips as substitutes to transit and bikesharing uses. A travel-time-based analysis further reveals that when choosing e-scooters over transit, travelers pay a price premium and save some travel time. The price premium is greater during the COVID-19 pandemic but the associated travel-time savings are smaller. This implies that public health considerations rather than time-cost tradeoffs are the main driver for many to choose e-scooters over transit during COVID. In addition, we find that e-scooters complement bikesharing and transit by providing services to underserved neighborhoods. A sizeable proportion (about 10 percent) of e-scooter trips are taken to connect with the rail services. Future research may combine the big-data-based analysis presented here with traditional methods to further shed light on the interactions between e-scooter services, bikesharing, and public transit.

cs.CY↗

arXiv2021

Analysis of The Ratio of $\ell_1$ and $\ell_2$ Norms in Compressed Sensing

We first propose a novel criterion that guarantees that an $s$-sparse signal is the local minimizer of the $\ell_1/\ell_2$ objective; our criterion is interpretable and useful in practice. We also give the first uniform recovery condition using a geometric characterization of the null space of the measurement matrix, and show that this condition is easily satisfied for a class of random matrices. We also present analysis on the robustness of the procedure when noise pollutes data. Numerical experiments are provided that compare $\ell_1/\ell_2$ with some other popular non-convex methods in compressed sensing. Finally, we propose a novel initialization approach to accelerate the numerical optimization procedure. We call this initialization approach \emph{support selection}, and we demonstrate that it empirically improves the performance of existing $\ell_1/\ell_2$ algorithms.

math.NA↗

arXiv2021

Infections Forecasting and Intervention Effect Evaluation for COVID-19 via a Data-Driven Markov Process and Heterogeneous Simulation

The Coronavirus Disease 2019 (COVID-19) pandemic has caused tremendous amount of deaths and a devastating impact on the economic development all over the world. Thus, it is paramount to control its further transmission, for which purpose it is necessary to find the mechanism of its transmission process and evaluate the effect of different control strategies. To deal with these issues, we describe the transmission of COVID-19 as an explosive Markov process with four parameters. The state transitions of the proposed Markov process can clearly disclose the terrible explosion and complex heterogeneity of COVID-19. Based on this, we further propose a simulation approach with heterogeneous infections. Experimentations show that our approach can closely track the real transmission process of COVID-19, disclose its transmission mechanism, and forecast the transmission under different non-drug intervention strategies. More importantly, our approach can helpfully develop effective strategies for controlling COVID-19 and appropriately compare their control effect in different countries/cities.

stat.AP↗

arXiv2020

Open Set Domain Adaptation by Extreme Value Theory

Common domain adaptation techniques assume that the source domain and the target domain share an identical label space, which is problematic since when target samples are unlabeled we have no knowledge on whether the two domains share the same label space. When this is not the case, the existing methods fail to perform well because the additional unknown classes are also matched with the source domain during adaptation. In this paper, we tackle the open set domain adaptation problem under the assumption that the source and the target label spaces only partially overlap, and the task becomes when the unknown classes exist, how to detect the target unknown classes and avoid aligning them with the source domain. We propose to utilize an instance-level reweighting strategy for domain adaptation where the weights indicate the likelihood of a sample belonging to known classes and to model the tail of the entropy distribution with Extreme Value Theory for unknown class detection. Experiments on conventional domain adaptation datasets show that the proposed method outperforms the state-of-the-art models.

stat.ML↗

arXiv2020

Concept Drift and Covariate Shift Detection Ensemble with Lagged Labels

In model serving, having one fixed model during the entire often life-long inference process is usually detrimental to model performance, as data distribution evolves over time, resulting in lack of reliability of the model trained on historical data. It is important to detect changes and retrain the model in time. The existing methods generally have three weaknesses: 1) using only classification error rate as signal, 2) assuming ground truth labels are immediately available after features from samples are received and 3) unable to decide what data to use to retrain the model when change occurs. We address the first problem by utilizing six different signals to capture a wide range of characteristics of data, and we address the second problem by allowing lag of labels, where labels of corresponding features are received after a lag in time. For the third problem, our proposed method automatically decides what data to use to retrain based on the signals. Extensive experiments on structured and unstructured data for different type of data changes establish that our method consistently outperforms the state-of-the-art methods by a large margin.

cs.AI↗

arXiv2020

Micromobility Trip Origin and Destination Inference Using General Bikeshare Feed Specification (GBFS) Data

Emerging micromobility services (e.g., e-scooters) have a great potential to enhance urban mobility but more knowledge on their usage patterns is needed. The General Bikeshare Feed Specification (GBFS) data are a possible source for examining micromobility trip patterns, but efforts are needed to infer trips from the GBFS data. Existing trip inference methods are usually based upon the assumption that the vehicle ID of a micromobility option (e-scooter or e-bike) does not change, and so they cannot deal with data with vehicle IDs that change over time. In this study, we propose a comprehensive package of algorithms to infer trip origins and destinations from GBFS data with different types of vehicle ID. We implement the algorithms in Washington DC by analyzing one-week (last week of February 2020) of GBFS data published by six vendors, and we evaluate the inference accuracy of the proposed algorithms by R-squared, mean absolute error, and sum absolute error. We find that the R-squared measure is larger than 0.9 and the MAE measure is smaller than 2 when the algorithms are evaluated with a 400m*400m grid, and the absolute errors are relatively larger in the downtown area. The accuracy of the trip-inference algorithms is sufficiently high for most practical applications.

cs.CY↗

arXiv2020

Consistency of archetypal analysis

Archetypal analysis is an unsupervised learning method that uses a convex polytope to summarize multivariate data. For fixed $k$, the method finds a convex polytope with $k$ vertices, called archetype points, such that the polytope is contained in the convex hull of the data and the mean squared distance between the data and the polytope is minimal. In this paper, we prove a consistency result that shows if the data is independently sampled from a probability measure with bounded support, then the archetype points converge to a solution of the continuum version of the problem, of which we identify and establish several properties. We also obtain the convergence rate of the optimal objective values under appropriate assumptions on the distribution. If the data is independently sampled from a distribution with unbounded support, we also prove a consistency result for a modified method that penalizes the dispersion of the archetype points. Our analysis is supported by detailed computational experiments of the archetype points for data sampled from the uniform distribution in a disk, the normal distribution, an annular distribution, and a Gaussian mixture model.

math.ST↗

arXiv2019

k-Nearest Neighbors by Means of Sequence to Sequence Deep Neural Networks and Memory Networks

k-Nearest Neighbors is one of the most fundamental but effective classification models. In this paper, we propose two families of models built on a sequence to sequence model and a memory network model to mimic the k-Nearest Neighbors model, which generate a sequence of labels, a sequence of out-of-sample feature vectors and a final label for classification, and thus they could also function as oversamplers. We also propose 'out-of-core' versions of our models which assume that only a small portion of data can be loaded into memory. Computational experiments show that our models on structured datasets outperform k-Nearest Neighbors, a feed-forward neural network, XGBoost, lightGBM, random forest and a memory network, due to the fact that our models must produce additional output and not just the label. On image and text datasets, the performance of our model is close to many state-of-the-art deep models. As an oversampler on imbalanced datasets, the sequence to sequence kNN model often outperforms Synthetic Minority Over-sampling Technique and Adaptive Synthetic Sampling.

cs.LG↗

arXiv2019

Open-Ended Visual Question Answering by Multi-Modal Domain Adaptation

We study the problem of visual question answering (VQA) in images by exploiting supervised domain adaptation, where there is a large amount of labeled data in the source domain but only limited labeled data in the target domain with the goal to train a good target model. A straightforward solution is to fine-tune a pre-trained source model by using those limited labeled target data, but it usually cannot work well due to the considerable difference between the data distributions of the source and target domains. Moreover, the availability of multiple modalities (i.e., images, questions and answers) in VQA poses further challenges to model the transferability between those different modalities. In this paper, we tackle the above issues by proposing a novel supervised multi-modal domain adaptation method for VQA to learn joint feature embeddings across different domains and modalities. Specifically, we align the data distributions of the source and target domains by considering all modalities together as well as separately for each individual modality. Based on the extensive experiments on the benchmark VQA 2.0 and VizWiz datasets for the realistic open-ended VQA task, we demonstrate that our proposed method outperforms the existing state-of-the-art approaches in this challenging domain adaptation setting for VQA.

cs.CV↗

arXiv2019

Automatic Ontology Learning from Domain-Specific Short Unstructured Text Data

Ontology learning is a critical task in industry, dealing with identifying and extracting concepts captured in text data such that these concepts can be used in different tasks, e.g. information retrieval. Ontology learning is non-trivial due to several reasons with limited amount of prior research work that automatically learns a domain specific ontology from data. In our work, we propose a two-stage classification system to automatically learn an ontology from unstructured text data. We first collect candidate concepts, which are classified into concepts and irrelevant collocates by our first classifier. The concepts from the first classifier are further classified by the second classifier into different concept types. The proposed system is deployed as a prototype at a company and its performance is validated by using complaint and repair verbatim data collected in automotive industry from different data sources.

cs.IR↗

arXiv2019

A setup for extreme-ultraviolet ultrafast angle-resolved photoelectron spectroscopy at 50-kHz repetition rate

Time- and angle-resolved photoelectron spectroscopy (trARPES) is a powerful method to track the ultrafast dynamics of quasiparticles and electronic bands in energy and momentum space. We present a setup for trARPES with 22.3 eV extreme-ultraviolet (XUV) femtosecond pulses at 50-kHz repetition rate, which enables fast data acquisition and access to dynamics across momentum space with high sensitivity. The design and operation of the XUV beamline, pump-probe setup, and UHV endstation are described in detail. By characterizing the effect of space-charge broadening, we determine an ultimate source-limited energy resolution of 60 meV, with typically 80-100 meV obtained at 1-2e10 photons/s probe flux on the sample. The instrument capabilities are demonstrated via both equilibrium and time-resolved ARPES studies of transition-metal dichalcogenides. The 50-kHz repetition rate enables sensitive measurements of quasiparticles at low excitation fluences in semiconducting MoSe$_2$, with an instrumental time resolution of 65 fs. Moreover, photo-induced phase transitions can be driven with the available pump fluence, as shown by charge density wave melting in 1T-TiSe$_2$. The high repetition-rate setup thus provides a versatile platform for sensitive XUV trARPES, from quenching of electronic phases down to the perturbative limit.

physics.ins-det↗