Search

Quantile regression is a powerful tool for learning the relationship between a response variable and a multivariate predictor while exploring heterogeneous effects. In this paper, we consider statistical inference for quantile regression with large-scale data in the "increasing dimension" regime. We provide a comprehensive and in-depth analysis of a convolution-type smoothing approach that achieves adequate approximation to computation and inference for quantile regression. This method, which we refer to as {\it{conquer}}, turns the non-differentiable quantile loss function into a twice-differentiable, convex and locally strongly convex surrogate, which admits a fast and scalable Barzilai-Borwein gradient-based algorithm to perform optimization, and multiplier bootstrap for statistical inference. Theoretically, we establish explicit non-asymptotic bounds on both estimation and Bahadur-Kiefer linearization errors, from which we show that the asymptotic normality of the conquer estimator holds under a weaker requirement on the number of the regressors than needed for conventional quantile regression. Moreover, we prove the validity of multiplier bootstrap confidence constructions. Our numerical studies confirm the conquer estimator as a practical and reliable approach to large-scale inference for quantile regression. Software implementing the methodology is available in the \texttt{R} package \texttt{conquer}., An R package conquer for fitting smoothed quantile regression is available in CRAN, https://cran.r-project.org/web/packages/conquer/index.html

When no single outcome is sufficient to capture the multidimensional impairments of a disease, investigators often rely on multiple outcomes for comprehensive assessment of global disease status. Methods for assessing covariate effects on global disease status include the composite outcome and global test procedures. One global test procedure is the O'Brien's rank-sum test, which combines information from multiple outcomes using a global rank-sum score. However, existing methods for the global rank-sum do not lend themselves to regression modeling. We consider sensible regression strategies for the global percentile outcome (GPO), under the transformed linear model and the monotonic index model. Posing minimal assumptions, we develop estimation and inference procedures that account for the special features of the GPO. Asymptotics are established using U-statistic and U-process techniques. We illustrate the practical utilities of the proposed methods via extensive simulations and application to a Parkinson's disease study.

Regularized quantile regression (QR) is a useful technique for analyzing heterogeneous data under potentially heavy-tailed error contamination in high dimensions. This paper provides a new analysis of the estimation/prediction error bounds of the global solution of $L_1$ -regularized QR (QR-LASSO) and the local solutions of nonconvex regularized QR (QR-NCP) when the number of covariates is greater than the sample size. Our results build upon and significantly generalize the earlier work in the literature. For certain heavy-tailed error distributions and a general class of design matrices, the least-squares-based LASSO cannot achieve the near-oracle rate derived under the normality assumption no matter the choice of the tuning parameter. In contrast, we establish that QR-LASSO achieves the near-oracle estimation error rate for a broad class of models under conditions weaker than those in the literature. For QR-NCP, we establish the novel results that all local optima within a feasible region have desirable estimation accuracy. Our analysis applies to not just the hard sparsity setting commonly used in the literature, but also to the soft sparsity setting which permits many small coefficients. Our approach relies on a unified characterization of the global/local solutions of regularized QR via subgradients using a generalized Karush–Kuhn–Tucker condition. The theory of the paper establishes a key property of the subdifferential of the quantile loss function in high dimensions, which is of independent interest for analyzing other high-dimensional nonsmooth problems.

Quantile treatment effects are often considered in a quantile regression framework to adjust for the effect of covariates. In this study, we focus on the problem of testing whether the treatment ef...

Real-Time Bidding (RTB) is an important mechanism in modern online advertising systems. Advertisers employ bidding strategies in RTB to optimize their advertising effects subject to various financial requirements, especially the return-on-investment (ROI) constraint. ROIs change non-monotonically during the sequential bidding process, and often induce a see-saw effect between constraint satisfaction and objective optimization. While some existing approaches show promising results in static or mildly changing ad markets, they fail to generalize to highly dynamic ad markets with ROI constraints, due to their inability to adaptively balance constraints and objectives amidst non-stationarity and partial observability. In this work, we specialize in ROI-Constrained Bidding in non-stationary markets. Based on a Partially Observable Constrained Markov Decision Process, our method exploits an indicator-augmented reward function free of extra trade-off parameters and develops a Curriculum-Guided Bayesian Reinforcement Learning (CBRL) framework to adaptively control the constraint-objective trade-off in non-stationary ad markets. Extensive experiments on a large-scale industrial dataset with two problem settings reveal that CBRL generalizes well in both in-distribution and out-of-distribution data regimes, and enjoys superior learning efficiency and stability., Comment: Accepted by SIGKDD 2022

We present a context aware object detection method based on a retrieve-and-transform scene layout model. Given an input image, our approach first retrieves a coarse scene layout from a codebook of typical layout templates. In order to handle large layout variations, we use a variant of the spatial transformer network to transform and refine the retrieved layout, resulting in a set of interpretable and semantically meaningful feature maps of object locations and scales. The above steps are implemented as a Layout Transfer Network which we integrate into Faster RCNN to allow for joint reasoning of object detection and scene layout estimation. Extensive experiments on three public datasets verified that our approach provides consistent performance improvements to the state-of-the-art object detection baselines on a variety of challenging tasks in the traffic surveillance and the autonomous driving domains., Comment: Paper accepted by the IEEE Transactions on Intelligent Transportation Systems

RNA sequencing data have been abundantly generated in biomedical research for biomarker discovery and other studies. Such data at the exon level are usually heavily tailed and correlated. Conventional statistical tests based on the mean or median difference for differential expression likely suffer from low power when the between-group difference occurs mostly in the upper or lower tail of the distribution of gene expression. We propose a tail-based test to make comparisons between groups in terms of a specific distribution area rather than a single location. The proposed test, which is derived from quantile regression, adjusts for covariates and accounts for within-sample dependence among the exons through a specified correlation structure. Through Monte Carlo simulation studies, we show that the proposed test is generally more powerful and robust in detecting differential expression than commonly used tests based on the mean or a single quantile. An application to TCGA lung adenocarcinoma data demonstrates the promise of the proposed method in terms of biomarker discovery.

Context Assessments of the duration of concussion recovery have primarily been limited to sport-related concussions and male contact sports. Furthermore, whereas durations of symptoms and return-to-activity (RTA) protocols encompass total recovery, the trajectory of each duration has not been examined separately. Objective To identify individual (eg, demographics, medical history), initial concussion injury (eg, symptoms), and external (eg, site) factors associated with symptom duration and RTA-protocol duration after concussion. Design Cohort study. Setting Three US military service academies. Patients or Other Participants A total of 10 604 cadets at participating US military service academies enrolled in the study and completed a baseline evaluation and up to 5 postinjury evaluations. A total of 726 cadets (451 men, 275 women) sustained concussions during the study period. Main Outcome Measure(s) Number of days from injury (1) until the participant became asymptomatic and (2) to complete the RTA protocol. Results Varsity athlete cadets took less time than nonvarsity cadets to become asymptomatic (hazard ratio [HR] = 1.75, 95% confidence interval = 1.38, 2.23). Cadets who reported less symptom severity on the Sport Concussion Assessment Tool, third edition (SCAT3), within 48 hours of concussion had 1.45 to 3.77 times shorter symptom-recovery durations than those with more symptom severity. Similar to symptom duration, varsity status was associated with a shorter RTA-protocol duration (HR = 1.74, 95% confidence interval = 1.34, 2.25), and less symptom severity on the SCAT3 was associated with a shorter RTA-protocol duration (HR range = 1.31 to 1.47). The academy that the cadet attended was associated with the RTA-protocol duration (P < .05). Conclusions The initial total number of symptoms reported and varsity athlete status were strongly associated with symptom and RTA-protocol durations. These findings suggested that external (varsity status and academy) and injury (symptom burden) factors influenced the time until RTA.

In this article, we propose a new technique for constructing confidence intervals for the mean of a noisy sequence with multiple change‐points. We use the weighted bootstrap to generalize the bootstrap aggregating or bagging estimator. A standard deviation formula for the bagging estimator is introduced, based on which smoothed confidence intervals are constructed. To further improve the performance of the smoothed interval for weak signals, we suggest a strategy of adaptively choosing between the percentile intervals and the smoothed intervals. A new intensity plot is proposed to visualize the pattern of the change‐points. We also propose a new change‐point estimator based on the intensity plot, which has superior performance in comparison with the state‐of‐the‐art segmentation methods. The finite sample performance of the confidence intervals and the change‐point estimator are evaluated through Monte Carlo studies and illustrated with a real data example.

Ultrasound spine imaging technique has been applied to the assessment of spine deformity. However, manual measurements of scoliotic angles on ultrasound images are time-consuming and heavily rely on raters experience. The objectives of this study are to construct a fully automatic framework based on Faster R-CNN for detecting vertebral lamina and to measure the fitting spinal curves from the detected lamina pairs. The framework consisted of two closely linked modules: 1) the lamina detector for identifying and locating each lamina pairs on ultrasound coronal images, and 2) the spinal curvature estimator for calculating the scoliotic angles based on the chain of detected lamina. Two hundred ultrasound images obtained from AIS patients were identified and used for the training and evaluation of the proposed method. The experimental results showed the 0.76 AP on the test set, and the Mean Absolute Difference (MAD) between automatic and manual measurement which was within the clinical acceptance error. Meanwhile the correlation between automatic measurement and Cobb angle from radiographs was 0.79. The results revealed that our proposed technique could provide accurate and reliable automatic curvature measurements on ultrasound spine images for spine deformities., Accepted by IUS2021

Censored quantile regression (CQR) has become a valuable tool to study the heterogeneous association between a possibly censored outcome and a set of covariates, yet computation and statistical inference for CQR have remained a challenge for large-scale data with many covariates. In this paper, we focus on a smoothed martingale-based sequential estimating equations approach, to which scalable gradient-based algorithms can be applied. Theoretically, we provide a unified analysis of the smoothed sequential estimator and its penalized counterpart in increasing dimensions. When the covariate dimension grows with the sample size at a sublinear rate, we establish the uniform convergence rate (over a range of quantile indexes) and provide a rigorous justification for the validity of a multiplier bootstrap procedure for inference. In high-dimensional sparse settings, our results considerably improve the existing work on CQR by relaxing an exponential term of sparsity. We also demonstrate the advantage of the smoothed CQR over existing methods with both simulated experiments and data applications.

Weakly supervised nuclei segmentation is a critical problem for pathological image analysis and greatly benefits the community due to the significant reduction of labeling cost. Adopting point annotations, previous methods mostly rely on less expressive representations for nuclei instances and thus have difficulty in handling crowded nuclei. In this paper, we propose to decouple weakly supervised semantic and instance segmentation in order to enable more effective subtask learning and to promote instance-aware representation learning. To achieve this, we design a modular deep network with two branches: a semantic proposal network and an instance encoding network, which are trained in a two-stage manner with an instance-sensitive loss. Empirical results show that our approach achieves the state-of-the-art performance on two public benchmarks of pathological images from different types of organs., Comment: Accepted by ISBI 2022 as Oral Presentation

Time-delay reservoir computing uses a nonlinear node associated with a feedback loop to construct a large number of virtual neurons in the neural network. The clock cycle of the computing network is usually synchronous with the delay time of the feedback loop, which substantially constrains the flexibility of hardware implementations. This work shows an asynchronous reservoir computing network based on a semiconductor laser with an optical feedback loop, where the clock cycle (20 ns) is considerably different to the delay time (77 ns). The performance of this asynchronous network is experimentally investigated under various operation conditions. It is proved that the asynchronous reservoir computing shows highly competitive performance on the prediction task of Santa Fe chaotic time series, in comparison with the synchronous counterparts.

Reconstructing 3D object from a single image (RGB or depth) is a fundamental problem in visual scene understanding and yet remains challenging due to its ill-posed nature and complexity in real-world scenes. To address those challenges, we adopt a primitive-based representation for 3D object, and propose a two-stage graph network for primitive-based 3D object estimation, which consists of a sequential proposal module and a graph reasoning module. Given a 2D image, our proposal module first generates a sequence of 3D primitives from input image with local feature attention. Then the graph reasoning module performs joint reasoning on a primitive graph to capture the global shape context for each primitive. Such a framework is capable of taking into account rich geometry and semantic constraints during 3D structure recovery, producing 3D objects with more coherent structure even under challenging viewing conditions. We train the entire graph neural network in a stage-wise strategy and evaluate it on three benchmarks: Pix3D, ModelNet and NYU Depth V2. Extensive experiments show that our approach outperforms the previous state of the arts with a considerable margin., Comment: Accepted by ACM MM'21

We introduce GNeRF, a framework to marry Generative Adversarial Networks (GAN) with Neural Radiance Field (NeRF) reconstruction for the complex scenarios with unknown and even randomly initialized camera poses. Recent NeRF-based advances have gained popularity for remarkable realistic novel view synthesis. However, most of them heavily rely on accurate camera poses estimation, while few recent methods can only optimize the unknown camera poses in roughly forward-facing scenes with relatively short camera trajectories and require rough camera poses initialization. Differently, our GNeRF only utilizes randomly initialized poses for complex outside-in scenarios. We propose a novel two-phases end-to-end framework. The first phase takes the use of GANs into the new realm for optimizing coarse camera poses and radiance fields jointly, while the second phase refines them with additional photometric loss. We overcome local minima using a hybrid and iterative optimization scheme. Extensive experiments on a variety of synthetic and natural scenes demonstrate the effectiveness of GNeRF. More impressively, our approach outperforms the baselines favorably in those scenes with repeated patterns or even low textures that are regarded as extremely challenging before., Comment: ICCV 2021 (Oral)

The mesoscale description of the subcellular organization informs about cellular mechanisms in disease state. However, applications of soft X-ray tomography (SXT), an important approach for characterizing organelle organization, are limited by labor-intensive manual segmentation. Here we report a pipeline for automated segmentation and systematic analysis of SXT tomograms. Our approach combines semantic and first-applied instance segmentation to produce separate organelle masks with high Dice and Recall indexes, followed by analysis of organelle localization based on the radial distribution function. We demonstrated this technique by investigating the organization of INS-1E pancreatic β-cell organization under different treatments at multiple time points. Consistent with a previous analysis of a similar dataset, our results revealed the impact of glucose stimulation on the localization and molecular density of insulin vesicles and mitochondria. This pipeline can be extended to SXT tomograms of any cell type to shed light on the subcellular rearrangements under different drug treatments.

In the present paper we investigate the predictive risk of possibly misspecified quantile regression functions. The in-sample risk is well-known to be an overly optimistic estimate of the predictive risk and we provide two relatively simple (asymptotic) characterizations of the associated bias, also called expected optimism. We propose estimates for the expected optimism and the predictive risk, and establish their uniform consistency under mild conditions. Our results hold for models of moderately growing size and allow the quantile function to be incorrectly specified. Empirical evidence from our estimates is encouraging as it compares favorably with cross-validation.

This article concerns the potential bias in statistical inference on treatment effects when a large number of covariates are present in a linear or partially linear model. While the estimation bias in an under-fitted model is well understood, we address a lesser-known bias that arises from an over-fitted model. The over-fitting bias can be eliminated through data splitting at the cost of statistical efficiency, and we show that smoothing over random data splits can be pursued to mitigate the efficiency loss. We also discuss some of the existing methods for debiased inference and provide insights into their intrinsic bias-variance trade-off, which leads to an improvement in bias controls. Under appropriate conditions, we show that the proposed estimators for the treatment effects are asymptotically normal and their variances can be well estimated. We discuss the pros and cons of various methods both theoretically and empirically, and show that the proposed methods are valuable options in post-selection inference. Supplementary materials for this article are available online.

Few-shot video classification aims to learn new video categories with only a few labeled examples, alleviating the burden of costly annotation in real-world applications. However, it is particularly challenging to learn a class-invariant spatial-temporal representation in such a setting. To address this, we propose a novel matching-based few-shot learning strategy for video sequences in this work. Our main idea is to introduce an implicit temporal alignment for a video pair, capable of estimating the similarity between them in an accurate and robust manner. Moreover, we design an effective context encoding module to incorporate spatial and feature channel context, resulting in better modeling of intra-class variations. To train our model, we develop a multi-task loss for learning video matching, leading to video features with better generalization. Extensive experimental results on two challenging benchmarks, show that our method outperforms the prior arts with a sizable margin on Something-Something-V2 and competitive results on Kinetics.

The value of hypothesis testing, and the frequent misinterpretation of p-values as a cornerstone of statistical methodology, continues to be debated. In 2019, the President of the American Statisti...

We address the problem of class incremental learning, which is a core step towards achieving adaptive vision intelligence. In particular, we consider the task setting of incremental learning with limited memory and aim to achieve better stability-plasticity trade-off. To this end, we propose a novel two-stage learning approach that utilizes a dynamically expandable representation for more effective incremental concept modeling. Specifically, at each incremental step, we freeze the previously learned representation and augment it with additional feature dimensions from a new learnable feature extractor. This enables us to integrate new visual concepts with retaining learned knowledge. We dynamically expand the representation according to the complexity of novel concepts by introducing a channel-level mask-based pruning strategy. Moreover, we introduce an auxiliary loss to encourage the model to learn diverse and discriminate features for novel concepts. We conduct extensive experiments on the three class incremental learning benchmarks and our method consistently outperforms other methods with a large margin., Comment: Accepted as an oral of CVPR2021

Phosphoproteomics integrating data-independent acquisition (DIA) has enabled deep phosphoproteome profiling with improved quantification reproducibility and accuracy compared to data-dependent acquisition (DDA)-based phosphoproteomics. DIA data mining heavily relies on a spectral library that in most cases is built on DDA analysis of the same sample. Construction of this project-specific DDA library impairs the analytical throughput, limits the proteome coverage, and increases the sample size for DIA phosphoproteomics. Herein we introduce a novel deep neural network, DeepPhospho, which conceptually differs from previous deep learning models to achieve accurate predictions of LC-MS/MS data for phosphopeptides. By leveraging in silico libraries generated by DeepPhospho, we established a new DIA workflow for phosphoproteome profiling which involves DIA data acquisition and data mining with DeepPhospho predicted libraries, thus circumventing the need of DDA library construction. Our DeepPhospho-empowered workflow substantially expanded the phosphoproteome coverage while maintaining high quantification performance, which led to the discovery of more signaling pathways and regulated kinases in an EGF signaling study than the DDA library-based approach. DeepPhospho is provided as a web server to facilitate user access to predictions and library generation.

Phosphoproteomics integrating data-independent acquisition (DIA) enables deep phosphoproteome profiling with improved quantification reproducibility and accuracy compared to data-dependent acquisition (DDA)-based phosphoproteomics. DIA data mining heavily relies on a spectral library that in most cases is built on DDA analysis of the same sample. Construction of this project-specific DDA library impairs the analytical throughput, limits the proteome coverage, and increases the sample size for DIA phosphoproteomics. Herein we introduce a deep neural network, DeepPhospho, which conceptually differs from previous deep learning models to achieve accurate predictions of LC-MS/MS data for phosphopeptides. By leveraging in silico libraries generated by DeepPhospho, we establish a DIA workflow for phosphoproteome profiling which involves DIA data acquisition and data mining with DeepPhospho predicted libraries, thus circumventing the need of DDA library construction. Our DeepPhospho-empowered workflow substantially expands the phosphoproteome coverage while maintaining high quantification performance, which leads to the discovery of more signaling pathways and regulated kinases in an EGF signaling study than the DDA library-based approach. DeepPhospho is provided as a web server as well as an offline app to facilitate user access to model training, predictions and library generation., The coverage and throughput of data-independent acquisition (DIA)-based phosphoproteomics is limited by its dependence on experimental spectral libraries. Here the authors develop a DIA workflow based on in silico spectral libraries generated by a novel deep neural network to expand phosphoproteome coverage.

Despite the recent success of deep neural networks, it remains challenging to effectively model the long-tail class distribution in visual recognition tasks. To address this problem, we first investigate the performance bottleneck of the two-stage learning framework via ablative study. Motivated by our discovery, we propose a unified distribution alignment strategy for long-tail visual recognition. Specifically, we develop an adaptive calibration function that enables us to adjust the classification scores for each data point. We then introduce a generalized re-weight method in the two-stage learning to balance the class prior, which provides a flexible and unified solution to diverse scenarios in visual recognition tasks. We validate our method by extensive experiments on four tasks, including image classification, semantic segmentation, object detection, and instance segmentation. Our approach achieves the state-of-the-art results across all four recognition tasks with a simple and unified framework. The code and models will be made publicly available at: https://github.com/Megvii-BaseDetection/DisAlign, Accepted by CVPR 2021

Time-delay reservoir computer (RC) based on semiconductor lasers provides a simple hardware implementation of the recurrent neural network. However, the data processing speed is limited by the length of the feedback loop. This work demonstrates a parallel RC scheme based on the wavelength division multiplexing (WDM) technique. This scheme is implemented on a Fabry-Perot quantum dot laser with multimode emission. It is shown that the four-channel WDM RC exhibits a better performance over the single-channel one, with the same number of virtual neurons. Meanwhile, the RC is accelerated by four times, owing to the shorter delay time. In addition, we show that the cross-gain saturation effect between the multimodes plays a crucial role on the RC performance.

In this paper, we look into the problem of disentangled representation learning and controllable image synthesis in a deep generative model. We develop an encoder-decoder architecture for a variant of the Variational Auto-Encoder (VAE) with two latent codes $z_{1}$ and $z_{2}$ . Our framework uses $z_{2}$ to capture specified factors of variation while $z_{1}$ captures the complementary factors of variation. To this end, we analyze the learning problem from the perspective of multivariate mutual information, derive optimizable lower bounds of the conditional mutual information in the image synthesis processes and incorporate them into the training objective. We validate our method empirically on the Color MNIST dataset and the CelebA dataset by showing controllable image syntheses. Our proposed paradigm is simple yet effective and is applicable to many situations, including those where there is not an explicit factorization of features available, or where the features are non-categorical.

Scene graph generation is an important visual understanding task with a broad range of vision applications. Despite recent tremendous progress, it remains challenging due to the intrinsic long-tailed class distribution and large intra-class variation. To address these issues, we introduce a novel confidence-aware bipartite graph neural network with adaptive message propagation mechanism for unbiased scene graph generation. In addition, we propose an efficient bi-level data resampling strategy to alleviate the imbalanced data distribution problem in training our graph network. Our approach achieves superior or competitive performance over previous methods on several challenging datasets, including Visual Genome, Open Images V4/V6, demonstrating its effectiveness and generality., Comment: Accepted by CVPR2021

Visual grounding, which aims to build a correspondence between visual objects and their language entities, plays a key role in cross-modal scene understanding. One promising and scalable strategy for learning visual grounding is to utilize weak supervision from only image-caption pairs. Previous methods typically rely on matching query phrases directly to a precomputed, fixed object candidate pool, which leads to inaccurate localization and ambiguous matching due to lack of semantic relation constraints. In our paper, we propose a novel context-aware weakly-supervised learning method that incorporates coarse-to-fine object refinement and entity relation modeling into a two-stage deep network, capable of producing more accurate object representation and matching. To effectively train our network, we introduce a self-taught regression loss for the proposal locations and a classification loss based on parsed entity relations. Extensive experiments on two public benchmarks Flickr30K Entities and ReferItGame demonstrate the efficacy of our weakly grounding framework. The results show that we outperform the previous methods by a considerable margin, achieving 59.27\% top-1 accuracy in Flickr30K Entities and 37.68\% in the ReferItGame dataset respectively (Code is available at https://github.com/youngfly11/ReIR-WeaklyGrounding.pytorch.git)., Accepted by CVPR2021

Incremental learning of semantic segmentation has emerged as a promising strategy for visual scene interpretation in the open- world setting. However, it remains challenging to acquire novel classes in an online fashion for the segmentation task, mainly due to its continuously-evolving semantic label space, partial pixelwise ground-truth annotations, and constrained data availability. To ad- dress this, we propose an incremental learning strategy that can fast adapt deep segmentation models without catastrophic forgetting, using a streaming input data with pixel annotations on the novel classes only. To this end, we develop a uni ed learning strategy based on the Expectation-Maximization (EM) framework, which integrates an iterative relabeling strategy that lls in the missing labels and a rehearsal-based incremental learning step that balances the stability-plasticity of the model. Moreover, our EM algorithm adopts an adaptive sampling method to select informative train- ing data and a class-balancing training strategy in the incremental model updates, both improving the e cacy of model learning. We validate our approach on the PASCAL VOC 2012 and ADE20K datasets, and the results demonstrate its superior performance over the existing incremental methods., Comment: Accepted by ACM MM'21

We consider a fixed-price mechanism design setting where a seller sells one item via a social network. Each buyer in the network has a valuation of the item independently derived from a given continuous distribution. Initially, the seller can only directly communicate with her neighbors and sells the item among them. In order to get a higher revenue, she needs more buyers to participate in the sale. One recent solution is to design dedicated mechanisms to incentivize buyers to invite their neighbors to join the sale, but they have relatively high time complexity and may evoke concern for privacy. We propose the very first fixed-price mechanism to achieve the same goal with less time complexity and better preservation of privacy. It improves the maximal expected revenue of the fixed-price mechanism without diffusion. Especially, when the valuation distribution is uniform on [0, 1], it guarantees a lower bound of the improvement.

Learning segmentation from noisy labels is an important task for medical image analysis due to the difficulty in acquiring high-quality annotations. Most existing methods neglect the pixel correlation and structural prior in segmentation, often producing noisy predictions around object boundaries. To address this, we adopt a superpixel representation and develop a robust iterative learning strategy that combines noise-aware training of segmentation network and noisy label refinement, both guided by the superpixels. This design enables us to exploit the structural constraints in segmentation labels and effectively mitigate the impact of label noise in learning. Experiments on two benchmarks show that our method outperforms recent state-of-the-art approaches, and achieves superior robustness in a wide range of label noises. Code is available at https://github.com/gaozhitong/SP_guided_Noisy_Label_Seg.

The task of skeleton-based action recognition remains a core challenge in human-centred scene understanding due to the multiple granularities and large variation in human motion. Existing approaches typically employ a single neural representation for different motion patterns, which has difficulty in capturing fine-grained action classes given limited training data. To address the aforementioned problems, we propose a novel multi-granular spatio-temporal graph network for skeleton-based action classification that jointly models the coarse- and fine-grained skeleton motion patterns. To this end, we develop a dual-head graph network consisting of two interleaved branches, which enables us to extract features at two spatio-temporal resolutions in an effective and efficient manner. Moreover, our network utilises a cross-head communication strategy to mutually enhance the representations of both heads. We conducted extensive experiments on three large-scale datasets, namely NTU RGB+D 60, NTU RGB+D 120, and Kinetics-Skeleton, and achieves the state-of-the-art performance on all the benchmarks, which validates the effectiveness of our method., Comment: Accepted by ACM MM'21

OBJECTIVE To examine the factor structure of the Sport Concussion Assessment Tool-5 (SCAT5) symptom scale in adolescents on their initial presentation to a concussion clinic within the typical recovery period after concussion (ie

We present a novel real-time line segment detection scheme called Line Graph Neural Network (LGNN). Existing approaches require a computationally expensive verification or postprocessing step. Our LGNN employs a deep convolutional neural network (DCNN) for proposing line segment directly, with a graph neural network (GNN) module for reasoning their connectivities. Specifically, LGNN exploits a new quadruplet representation for each line segment where the GNN module takes the predicted candidates as vertexes and constructs a sparse graph to enforce structural context. Compared with the state-of-the-art, LGNN achieves near real-time performance without compromising accuracy. LGNN further enables time-sensitive 3D applications. When a 3D point cloud is accessible, we present a multi-modal line segment classification technique for extracting a 3D wireframe of the environment robustly and efficiently., 9 pages, 7 figures

As a discipline that deals with many aspects of data, statistics is a critical pillar in the rapidly evolving landscape of data science. The increasingly vital role of data, especially big data, in many applications, presents the field of statistics with unparalleled challenges and exciting opportunities. Statistics plays a pivotal role in data science by assisting with the use of data and decision making in the face of uncertainty. In this article, we present 10 research areas that could make statistics and data science more impactful on science and society. Focusing on these areas will help better transform data into knowledge, actionable insights and deliverables, and promote more collaboration with computer and other quantitative scientists and domain scientists.

Background Concussion, or mild traumatic brain injury, is a major public health concern affecting 42 million individuals globally each year. However, little is known regarding concussion risk factors across all concussion settings as most concussion research has focused on only sport-related or military-related concussive injuries. Methods The current study is part of the Concussion, Assessment, Research, and Education (CARE) Consortium, a multi-site investigation on the natural history of concussion. Cadets at three participating service academies completed annual baseline assessments, which included demographics, medical history, and concussion history, along with the Sport Concussion Assessment Tool (SCAT) symptom checklist and Brief Symptom Inventory (BSI-18). Clinical and research staff recorded the date and injury setting at time of concussion. Generalized mixed models estimated concussion risk with service academy as a random effect. Since concussion was a rare event, the odds ratios were assumed to approximate relative risk. Results Beginning in 2014, 10,604 (n = 2421, 22.83% female) cadets enrolled over 3 years. A total of 738 (6.96%) cadets experienced a concussion, 301 (2.84%) concussed cadets were female. Female sex and previous concussion were the most consistent estimators of concussion risk across all concussion settings. Compared to males, females had 2.02 (95% CI: 1.70–2.40) times the risk of a concussion regardless of injury setting, and greater relative risk when the concussion occurred during sport (Odds Ratio (OR): 1.38 95% CI: 1.07–1.78). Previous concussion was associated with 1.98 (95% CI: 1.65–2.37) times increased risk for any incident concussion, and the magnitude was relatively stable across all concussion settings (OR: 1.73 to 2.01). Freshman status was also associated with increased overall concussion risk, but was driven by increased risk for academy training-related concussions (OR: 8.17 95% CI: 5.87–11.37). Medical history of headaches in the past 3 months, diagnosed ADD/ADHD, and BSI-18 Somatization symptoms increased overall concussion risk. Conclusions Various demographic and medical history factors are associated with increased concussion risk. While certain factors (e.g. sex and previous concussion) are consistently associated with increased concussion risk, regardless of concussion injury setting, other factors significantly influence concussion risk within specific injury settings. Further research is required to determine whether these risk factors may aid in concussion risk reduction or prevention. Electronic supplementary material The online version of this article (10.1186/s40621-018-0178-3) contains supplementary material, which is available to authorized users.