Search

[EN] 1. Description of methods used for collection/generation of data: Diffraction (electron, synchrotron X-ray, neutron). Infrared spectroscopy multinuclear MAS NMR gas adsorption electron paramagnetic resonance DFT calculations Catalytic reaction 2. Methods for processing the data: Those needed depending on the type of data (SHELX, Rietveld refinement, fast Fourier Transform...). Please, check the paper.

The low-rank quaternion matrix approximation has been successfully applied in many applications involving signal processing and color image processing. However, the cost of quaternion models for generating low-rank quaternion matrix approximation is sometimes considerable due to the computation of the quaternion singular value decomposition (QSVD), which limits their application to real large-scale data. To address this deficiency, an efficient quaternion matrix CUR (QMCUR) method for low-rank approximation is suggested, which provides significant acceleration in color image processing. We first explore the QMCUR approximation method, which uses actual columns and rows of the given quaternion matrix, instead of the costly QSVD. Additionally, two different sampling strategies are used to sample the above-selected columns and rows. Then, the perturbation analysis is performed on the QMCUR approximation of noisy versions of low-rank quaternion matrices. Extensive experiments on both synthetic and real data further reveal the superiority of the proposed algorithm compared with other algorithms for getting low-rank approximation, in terms of both efficiency and accuracy.

Low-Light Image Enhancement (LLIE) task tends to restore the details and visual information from corrupted low-light images. Most existing methods learn the mapping function between low/normal-light images by Deep Neural Networks (DNNs) on sRGB and HSV color space. Nevertheless, enhancement involves amplifying image signals, and applying these color spaces to low-light images with a low signal-to-noise ratio can introduce sensitivity and instability into the enhancement process. Consequently, this results in the presence of color artifacts and brightness artifacts in the enhanced images. To alleviate this problem, we propose a novel trainable color space, named Horizontal/Vertical-Intensity (HVI). It not only decouples brightness and color from RGB channels to mitigate the instability during enhancement but also adapts to low-light images in different illumination ranges due to the trainable parameters. Further, we design a novel Color and Intensity Decoupling Network (CIDNet) with two branches dedicated to processing the decoupled image brightness and color in the HVI space. Within CIDNet, we introduce the Lightweight Cross-Attention (LCA) module to facilitate interaction between image structure and content information in both branches, while also suppressing noise in low-light images. Finally, we conducted 22 quantitative and qualitative experiments to show that the proposed CIDNet outperforms the state-of-the-art methods on 11 datasets. The code is available at https://github.com/Fediory/HVI-CIDNet.

Discovering causal relationships from observational data, particularly in the presence of latent variables, poses a challenging problem. While current local structure learning methods have proven effective and efficient when the focus lies solely on the local relationships of a target variable, they operate under the assumption of causal sufficiency. This assumption implies that all the common causes of the measured variables are observed, leaving no room for latent variables. Such a premise can be easily violated in various real-world applications, resulting in inaccurate structures that may adversely impact downstream tasks. In light of this, our paper delves into the primary investigation of locally identifying potential parents and children of a target from observational data that may include latent variables. Specifically, we harness the causal information from m-separation and V-structures to derive theoretical consistency results, effectively bridging the gap between global and local structure learning. Together with the newly developed stop rules, we present a principled method for determining whether a variable is a direct cause or effect of a target. Further, we theoretically demonstrate the correctness of our approach under the standard causal Markov and faithfulness conditions, with infinite samples. Experimental results on both synthetic and real-world data validate the effectiveness and efficiency of our approach.

Weakly supervised video anomaly detection (WSVAD) is a challenging task. Generating fine-grained pseudo-labels based on weak-label and then self-training a classifier is currently a promising solution. However, since the existing methods use only RGB visual modality and the utilization of category text information is neglected, thus limiting the generation of more accurate pseudo-labels and affecting the performance of self-training. Inspired by the manual labeling process based on the event description, in this paper, we propose a novel pseudo-label generation and self-training framework based on Text Prompt with Normality Guidance (TPWNG) for WSVAD. Our idea is to transfer the rich language-visual knowledge of the contrastive language-image pre-training (CLIP) model for aligning the video event description text and corresponding video frames to generate pseudo-labels. Specifically, We first fine-tune the CLIP for domain adaptation by designing two ranking losses and a distributional inconsistency loss. Further, we propose a learnable text prompt mechanism with the assist of a normality visual prompt to further improve the matching accuracy of video event description text and video frames. Then, we design a pseudo-label generation module based on the normality guidance to infer reliable frame-level pseudo-labels. Finally, we introduce a temporal context self-adaptive learning module to learn the temporal dependencies of different video events more flexibly and accurately. Extensive experiments show that our method achieves state-of-the-art performance on two benchmark datasets, UCF-Crime and XD-Viole, Comment: Accepted to CVPR2024

Deep neural networks (DNNs) are known to be sensitive to adversarial input perturbations, leading to a reduction in either prediction accuracy or individual fairness. To jointly characterize the susceptibility of prediction accuracy and individual fairness to adversarial perturbations, we introduce a novel robustness definition termed robust accurate fairness. Informally, robust accurate fairness requires that predictions for an instance and its similar counterparts consistently align with the ground truth when subjected to input perturbations. We propose an adversarial attack approach dubbed RAFair to expose false or biased adversarial defects in DNN, which either deceive accuracy or compromise individual fairness. Then, we show that such adversarial instances can be effectively addressed by carefully designed benign perturbations, correcting their predictions to be accurate and fair. Our work explores the double-edged sword of input perturbations to robust accurate fairness in DNN and the potential of using benign perturbations to correct adversarial instances.

Recommender systems learn personalized user preferences from user feedback like clicks. However, user feedback is usually biased towards partially observed interests, leaving many users' hidden interests unexplored. Existing approaches typically mitigate the bias, increase recommendation diversity, or use bandit algorithms to balance exploration-exploitation trade-offs. Nevertheless, they fail to consider the potential rewards of recommending different categories of items and lack the global scheduling of allocating top-N recommendations to categories, leading to suboptimal exploration. In this work, we propose an Uplift model-based Recommender (UpliftRec) framework, which regards top-N recommendation as a treatment optimization problem. UpliftRec estimates the treatment effects, i.e., the click-through rate (CTR) under different category exposure ratios, by using observational user feedback. UpliftRec calculates group-level treatment effects to discover users' hidden interests with high CTR rewards and leverages inverse propensity weighting to alleviate confounder bias. Thereafter, UpliftRec adopts a dynamic programming method to calculate the optimal treatment for overall CTR maximization. We implement UpliftRec on different backend models and conduct extensive experiments on three datasets. The empirical results validate the effectiveness of UpliftRec in discovering users' hidden interests while achieving superior recommendation accuracy., Comment: Accepted to SIGIR 2024

Empirical researchers and decision-makers spanning various domains frequently seek profound insights into the long-term impacts of interventions. While the significance of long-term outcomes is undeniable, an overemphasis on them may inadvertently overshadow short-term gains. Motivated by this, this paper formalizes a new framework for learning the optimal policy that effectively balances both long-term and short-term rewards, where some long-term outcomes are allowed to be missing. In particular, we first present the identifiability of both rewards under mild assumptions. Next, we deduce the semiparametric efficiency bounds, along with the consistency and asymptotic normality of their estimators. We also reveal that short-term outcomes, if associated, contribute to improving the estimator of the long-term reward. Based on the proposed estimators, we develop a principled policy learning approach and further derive the convergence rates of regret and estimation errors associated with the learned policy. Extensive experiments are conducted to validate the effectiveness of the proposed method, demonstrating its practical applicability.

Selection bias in recommender system arises from the recommendation process of system filtering and the interactive process of user selection. Many previous studies have focused on addressing selection bias to achieve unbiased learning of the prediction model, but ignore the fact that potential outcomes for a given user-item pair may vary with the treatments assigned to other user-item pairs, named neighborhood effect. To fill the gap, this paper formally formulates the neighborhood effect as an interference problem from the perspective of causal inference and introduces a treatment representation to capture the neighborhood effect. On this basis, we propose a novel ideal loss that can be used to deal with selection bias in the presence of neighborhood effect. We further develop two new estimators for estimating the proposed ideal loss. We theoretically establish the connection between the proposed and previous debiasing methods ignoring the neighborhood effect, showing that the proposed methods can achieve unbiased learning when both selection bias and neighborhood effect are present, while the existing methods are biased. Extensive semi-synthetic and real-world experiments are conducted to demonstrate the effectiveness of the proposed methods., Comment: ICLR 24

Debiased collaborative filtering aims to learn an unbiased prediction model by removing different biases in observational datasets. To solve this problem, one of the simple and effective methods is based on the propensity score, which adjusts the observational sample distribution to the target one by reweighting observed instances. Ideally, propensity scores should be learned with causal balancing constraints. However, existing methods usually ignore such constraints or implement them with unreasonable approximations, which may affect the accuracy of the learned propensity scores. To bridge this gap, in this paper, we first analyze the gaps between the causal balancing requirements and existing methods such as learning the propensity with cross-entropy loss or manually selecting functions to balance. Inspired by these gaps, we propose to approximate the balancing functions in reproducing kernel Hilbert space and demonstrate that, based on the universal property and representer theorem of kernel functions, the causal balancing constraints can be better satisfied. Meanwhile, we propose an algorithm that adaptively balances the kernel function and theoretically analyze the generalization error bound of our methods. We conduct extensive experiments to demonstrate the effectiveness of our methods, and to promote this research direction, we have released our project at https://github.com/haoxuanli-pku/ICLR24-Kernel-Balancing., Comment: ICLR 24 Spotlight

In real-world scenarios, images captured often suffer from blurring, noise, and other forms of image degradation, and due to sensor limitations, people usually can only obtain low dynamic range images. To achieve high-quality images, researchers have attempted various image restoration and enhancement operations on photographs, including denoising, deblurring, and high dynamic range imaging. However, merely performing a single type of image enhancement still cannot yield satisfactory images. In this paper, to deal with the challenge above, we propose the Composite Refinement Network (CRNet) to address this issue using multiple exposure images. By fully integrating information-rich multiple exposure inputs, CRNet can perform unified image restoration and enhancement. To improve the quality of image details, CRNet explicitly separates and strengthens high and low-frequency information through pooling layers, using specially designed Multi-Branch Blocks for effective fusion of these frequencies. To increase the receptive field and fully integrate input features, CRNet employs the High-Frequency Enhancement Module, which includes large kernel convolutions and an inverted bottleneck ConvFFN. Our model secured third place in the first track of the Bracketing Image Restoration and Enhancement Challenge, surpassing previous SOTA models in both testing metrics and visual quality., Comment: This paper is accepted by CVPR2024 Workshop, Code: https://github.com/CalvinYang0/CRNet

This paper reviews the NTIRE 2024 low light image enhancement challenge, highlighting the proposed solutions and results. The aim of this challenge is to discover an effective network design or solution capable of generating brighter, clearer, and visually appealing results when dealing with a variety of conditions, including ultra-high resolution (4K and beyond), non-uniform illumination, backlighting, extreme darkness, and night scenes. A notable total of 428 participants registered for the challenge, with 22 teams ultimately making valid submissions. This paper meticulously evaluates the state-of-the-art advancements in enhancing low-light images, reflecting the significant progress and creativity in this field., Comment: NTIRE 2024 Challenge Report

In real-world scenarios, due to a series of image degradations, obtaining high-quality, clear content photos is challenging. While significant progress has been made in synthesizing high-quality images, previous methods for image restoration and enhancement often overlooked the characteristics of different degradations. They applied the same structure to address various types of degradation, resulting in less-than-ideal restoration outcomes. Inspired by the notion that high/low frequency information is applicable to different degradations, we introduce HLNet, a Bracketing Image Restoration and Enhancement method based on high-low frequency decomposition. Specifically, we employ two modules for feature extraction: shared weight modules and non-shared weight modules. In the shared weight modules, we use SCConv to extract common features from different degradations. In the non-shared weight modules, we introduce the High-Low Frequency Decomposition Block (HLFDB), which employs different methods to handle high-low frequency information, enabling the model to address different degradations more effectively. Compared to other networks, our method takes into account the characteristics of different degradations, thus achieving higher-quality image restoration., Comment: This paper is accepted by CVPR 2024 Workshop, code: https://github.com/chengeng0613/HLNet

Telomeres replicated by leading-strand synthesis lack the 3 overhang required for telomere protection. Surprisingly, resection of these blunt telomeres is initiated by the telomere-specific 5 exonuclease Apollo rather than the Mre11-Rad50-Nbs1 (MRN) complex, the nuclease that acts at DNA breaks. Without Apollo, leading-end telomeres undergo fusion, which, as demonstrated here, is mediated by alternative end joining. Here, we show that DNA-PK and TRF2 coordinate the repression of MRN at blunt mouse telomeres. DNA-PK represses an MRN-dependent long-range resection, while the endonuclease activity of MRN-CtIP, which could cleave DNA-PK off of blunt telomere ends, is inhibited in vitro and in vivo by the iDDR of TRF2. AlphaFold-Multimer predicts a conserved association of the iDDR with Rad50, potentially interfering with CtIP binding and MRN endonuclease activation. We propose that repression of MRN-mediated resection is a conserved aspect of telomere maintenance and represents an ancient feature of DNA-PK and the iDDR., Funding Agencies|Cancerfonden [CAN 2018/493]; Vetenskapsradet [AR-MH 2018-03215]; NIH [R01GM138548, R35 CA210036, AG016642]; Knut and Alice Wallenberg Foundation; Lions forskningsfond [LiU-2022-01245]

Herein, we report on the discovery and development of novel cascade N-N bond forming reactions for the synthesis of rare indazole acetic acid scaffolds. This approach allows for convenient synthesis of three distinct indazole acetic acid derivatives (unsubstituted, hydroxy, and alkoxy) by heating 3-amino-3-(2-nitroaryl)propanoic acids with an appropriate nucleophile/solvent under basic conditions. The reaction tolerates a range of functional groups and electronic effects and, in total, 23 novel indazole acetic acids were synthesized and characterized. This work offers a valuable strategy for the synthesis of useful scaffolds for drug discovery programs.

This study attempts to address a challenge regarding the extraction of building information, which is one of the fundamental tasks that needs to be addressed in the construction domain. Current technologies, such as relational databases, have difficulty in efficiently and effectively managing and querying the interconnected building information with full of hidden relationships. To address this problem, this study adopted the graph-theory-based graph database technology to reveal hidden relationships within building information. A model-driven approach was developed to enable a full conversion of Industry Foundation Classes data into labeled property graph, which is referred to as IFC-Graph. The result shows that IFC-Graph can represent interconnected building information and reveal hidden relationships, supporting effective and efficient building information access and query. This study can benefit a vast number of future studies in the area of building information query by improving its accessibility and queryability.

Ting-Ting Wu,1,* Yi Qun Jiang,2,* Bang-Feng Zhao,1,* Feng-Li Si,1 Peng Wu,1 Huan-Ying Wang,1 Chun-Feng Sheng,1 Xun Xu,1 Fan Li,1 Jing Zhang3 1Songjiang Hospital Affiliated to Shanghai Jiao Tong University School of Medicine (Preparatory Stage), Shanghai, People’s Republic of China; 2General Practice/International Medical Care Center, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, People’s Republic of China; 3Department of Pulmonary and Critical Care Medicine, Zhongshan Hospital, Shanghai Medical College, Fudan University, Shanghai, People’s Republic of China*These authors contributed equally to this workCorrespondence: Fan Li; Jing Zhang, Email lifanking2015@163.com; jingatlas@outlook.comObjective: To evaluate the real-world situation for the management of chronic obstructive pulmonary disease (COPD) and poorly controlled disease risk factors in the Chinese community.Methods: This retrospective multicentre study analysed data from COPDMICand MICHC in Shanghai Songjiang District, Shanghai, China. The differences in COPD Assessment Test (CAT), the modified Medical Research Council (mMRC) dyspnea scale, and the number of emergency cases, emergency visits, inpatient cases, and hospitalisations from January 2018 to December 2020 were analysed. The impact of coronavirus disease 2019 (COVID-19) on COPD management was also assessed.Results: For 2020 versus 2018, analysis of 468 COPD cases from COPDMIC matched with MICHC data showed significantly more patients with improved mMRC grades, significantly fewer emergency cases and emergency visits, and significantly fewer hospitalisation cases and hospitalisations. Differences in the number of emergency visits and hospitalisations per capita were statistically significant. Compared to GOLD 3– 4, GOLD 1– 2 patients showed significant improvements in CAT score, mMRC grade, the number of emergency visits and hospitalisations per capita. Treatment adher

Jamming signals can jeopardize the operation of GNSS receivers until denying its operation. Given their ubiquity, jamming mitigation and localization techniques are of crucial importance, for which jammer classification is of help. Data-driven models have been proven useful in detecting these threats, while their training using crowdsourced data still poses challenges when it comes to private data sharing. This article investigates the use of federated learning to train jamming signal classifiers locally on each device, with model updates aggregated and averaged at the central server. This allows for privacy-preserving training procedures that do not require centralized data storage or access to client local data. The used framework FedAvg is assessed on a dataset consisting of spectrogram images of simulated interfered GNSS signal. Six different jammer types are effectively classified with comparable results to a fully centralized solution that requires vast amounts of data communication and involves privacy-preserving concerns.

Deep neural networks (DNNs) often face challenges due to their vulnerability to various adversarial perturbations, including false perturbations that undermine prediction accuracy and biased perturbations that cause biased predictions for similar inputs. This paper introduces a novel approach, RobustFair, to evaluate the accurate fairness of DNNs when subjected to these false or biased perturbations. RobustFair employs the notion of the fairness confusion matrix induced in accurate fairness to identify the crucial input features for perturbations. This matrix categorizes predictions as true fair, true biased, false fair, and false biased, and the perturbations guided by it can produce a dual impact on instances and their similar counterparts to either undermine prediction accuracy (robustness) or cause biased predictions (individual fairness). RobustFair then infers the ground truth of these generated adversarial instances based on their loss function values approximated by the total derivative. To leverage the generated instances for trustworthiness improvement, RobustFair further proposes a data augmentation strategy to prioritize adversarial instances resembling the original training set, for data augmentation and model retraining. Notably, RobustFair excels at detecting intertwined issues of robustness and individual fairness, which are frequently overlooked in standard robustness and individual fairness evaluations. This capability empowers RobustFair to enhance both robustness and individual fairness evaluations by concurrently identifying defects in either domain. Empirical case studies and quantile regression analyses on benchmark datasets demonstrate the effectiveness of the fairness confusion matrix guided perturbation for false or biased adversarial instance generation.

Recommender systems are seen as an effective tool to address information overload, but it is widely known that the presence of various biases makes direct training on large-scale observational data result in sub-optimal prediction performance. In contrast, unbiased ratings obtained from randomized controlled trials or A/B tests are considered to be the golden standard, but are costly and small in scale in reality. To exploit both types of data, recent works proposed to use unbiased ratings to correct the parameters of the propensity or imputation models trained on the biased dataset. However, the existing methods fail to obtain accurate predictions in the presence of unobserved confounding or model misspecification. In this paper, we propose a theoretically guaranteed model-agnostic balancing approach that can be applied to any existing debiasing method with the aim of combating unobserved confounding and model misspecification. The proposed approach makes full use of unbiased data by alternatively correcting model parameters learned with biased data, and adaptively learning balance coefficients of biased samples for further debiasing. Extensive real-world experiments are conducted along with the deployment of our proposal on four representative debiasing methods to demonstrate the effectiveness., Comment: Accepted Paper in WWW'23

Video anomaly detection (VAD) is a significant computer vision problem. Existing deep neural network (DNN) based VAD methods mostly follow the route of frame reconstruction or frame prediction. However, the lack of mining and learning of higher-level visual features and temporal context relationships in videos limits the further performance of these two approaches. Inspired by video codec theory, we introduce a brand-new VAD paradigm to break through these limitations: First, we propose a new task of video event restoration based on keyframes. Encouraging DNN to infer missing multiple frames based on video keyframes so as to restore a video event, which can more effectively motivate DNN to mine and learn potential higher-level visual features and comprehensive temporal context relationships in the video. To this end, we propose a novel U-shaped Swin Transformer Network with Dual Skip Connections (USTN-DSC) for video event restoration, where a cross-attention and a temporal upsampling residual skip connection are introduced to further assist in restoring complex static and dynamic motion object features in the video. In addition, we propose a simple and effective adjacent frame difference loss to constrain the motion consistency of the video sequence. Extensive experiments on benchmarks demonstrate that USTN-DSC outperforms most existing methods, validating the effectiveness of our method., Comment: Accepted by CVPR 2023

Classification is an important task in the field of machine learning. Most classifiers based on offline learning are invalid for open data streams. In contrast, incremental learning is feasible for continuous data. This paper presents the Incremental Bayesian Classifier “Incremental_BC”, which continuously updates the probabilistic information according to each new training sample via recursive calculation. Further, the Incremental_BC is improved to deal with the flowing data whose distribution and property evolve with time, i.e., the concept drift. The effectiveness of the proposed methods has been verified by the results of simulation tests on benchmark data sets.

Human papillomavirus (HPV) integration and high expression of HPV oncogenes (E6 and E7) are important mechanisms for HPV carcinogenesis in cervical cancer. However, the relationship between HPV integration and HPV E6 spliced transcripts, as well as the underlying mechanisms of HPV integration in carcinogenesis after HPV E6 splicing remains unclear. We analyzed HPV-coiled-coil domain containing 106 (CCDC106) integration samples to characterize the roles of HPV integration, E6 spliceosome I (E6{*}I), and high CCDC106 expression in cervical carcinogenesis. We found that E6 was alternatively spliced into the E6{*}I transcript in HPV-CCDC016 integration samples with low p53 expression, in contrast to the role of E6{*}I in preventing p53 degradation in cervical cancer cells. In addition, CCDC106 was highly expressed after HPV-CCDC106 integration, and interacted with p53, resulting in p53 degradation and cervical cancer cell progression in vitro and in vivo. Importantly, when E6{*}I was highly expressed in cervical cancer cells, overexpression of CCDC106 independently degraded p53 and promoted cervical cancer cell progression. In this study, we explored the underlying mechanisms of HPV-CCDC106 integration in HPV carcinogenesis after HPV E6 splicing, which should provide insight into host genome dysregulation in cervical carcinogenesis.

Video anomaly detection (VAD) with weak supervision has achieved remarkable performance in utilizing video-level labels to discriminate whether a video frame is normal or abnormal. However, current approaches are inherently limited to a closed-set setting and may struggle in open-world applications where there can be anomaly categories in the test data unseen during training. A few recent studies attempt to tackle a more realistic setting, open-set VAD, which aims to detect unseen anomalies given seen anomalies and normal videos. However, such a setting focuses on predicting frame anomaly scores, having no ability to recognize the specific categories of anomalies, despite the fact that this ability is essential for building more informed video surveillance systems. This paper takes a step further and explores open-vocabulary video anomaly detection (OVVAD), in which we aim to leverage pre-trained large models to detect and categorize seen and unseen anomalies. To this end, we propose a model that decouples OVVAD into two mutually complementary tasks -- class-agnostic detection and class-specific classification -- and jointly optimizes both tasks. Particularly, we devise a semantic knowledge injection module to introduce semantic knowledge from large language models for the detection task, and design a novel anomaly synthesis module to generate pseudo unseen anomaly videos with the help of large vision generation models for the classification task. These semantic knowledge and synthesis anomalies substantially extend our model's capability in detecting and categorizing a variety of seen and unseen anomalies. Extensive experiments on three widely-used benchmarks demonstrate our model achieves state-of-the-art performance on OVVAD task., Comment: Accepted to CVPR2024

Class imbalance exists in many classification problems, and since the data is designed for accuracy, imbalance in data classes can lead to classification challenges with a few classes having higher misclassification costs. The Backblaze dataset, a widely used dataset related to hard discs, has a small amount of failure data and a large amount of health data, which exhibits a serious class imbalance. This paper provides a comprehensive overview of research in the field of imbalanced data classification. The discussion is organized into three main aspects: data-level methods, algorithmic-level methods, and hybrid methods. For each type of method, we summarize and analyze the existing problems, algorithmic ideas, strengths, and weaknesses. Additionally, the challenges of unbalanced data classification are discussed, along with strategies to address them. It is convenient for researchers to choose the appropriate method according to their needs.

Data class imbalance is a common problem in classification problems, where minority class samples are often more important and more costly to misclassify in a classification task. Therefore, it is very important to solve the data class imbalance classification problem. The SMART dataset exhibits an evident class imbalance, comprising a substantial quantity of healthy samples and a comparatively limited number of defective samples. This dataset serves as a reliable indicator of the disc's health status. In this paper, we obtain the best balanced disk SMART dataset for a specific classification model by mixing and integrating the data synthesised by multivariate generative adversarial networks (GAN) to balance the disk SMART dataset at the data level; and combine it with genetic algorithms to obtain higher disk fault classification prediction accuracy on a specific classification model.

Large scale data storage is susceptible to failure. As disks are damaged and replaced, traditional machine learning models, which rely on historical data to make predictions, struggle to accurately predict disk failures. This paper presents a novel method for predicting disk failures by leveraging multi-layer domain adaptive learning techniques. First, disk data with numerous faults is selected as the source domain, and disk data with fewer faults is selected as the target domain. A training of the feature extraction network is performed with the selected origin and destination domains. The contrast between the two domains facilitates the transfer of diagnostic knowledge from the domain of source and target. According to the experimental findings, it has been demonstrated that the proposed technique can generate a reliable prediction model and improve the ability to predict failures on disk data with few failure samples.

To address the problem of insufficient failure data generated by disks and the imbalance between the number of normal and failure data. The existing Conditional Tabular Generative Adversarial Networks (CTGAN) deep learning methods have been proven to be effective in solving imbalance disk failure data. But CTGAN cannot learn the internal information of disk failure data very well. In this paper, a fault diagnosis method based on improved CTGAN, a classifier for specific category discrimination is added and a discriminator generate adversarial network based on residual network is proposed. We named it Residual Conditional Tabular Generative Adversarial Networks (RCTGAN). Firstly, to enhance the stability of system a residual network is utilized. RCTGAN uses a small amount of real failure data to synthesize fake fault data; Then, the synthesized data is mixed with the real data to balance the amount of normal and failure data; Finally, four classifier (multilayer perceptron, support vector machine, decision tree, random forest) models are trained using the balanced data set, and the performance of the models is evaluated using G-mean. The experimental results show that the data synthesized by the RCTGAN can further improve the fault diagnosis accuracy of the classifier.

In the domain of video surveillance, describing the behavior of each individual within the video is becoming increasingly essential, especially in complex scenarios with multiple individuals present. This is because describing each individual's behavior provides more detailed situational analysis, enabling accurate assessment and response to potential risks, ensuring the safety and harmony of public places. Currently, video-level captioning datasets cannot provide fine-grained descriptions for each individual's specific behavior. However, mere descriptions at the video-level fail to provide an in-depth interpretation of individual behaviors, making it challenging to accurately determine the specific identity of each individual. To address this challenge, we construct a human-centric video surveillance captioning dataset, which provides detailed descriptions of the dynamic behaviors of 7,820 individuals. Specifically, we have labeled several aspects of each person, such as location, clothing, and interactions with other elements in the scene, and these people are distributed across 1,012 videos. Based on this dataset, we can link individuals to their respective behaviors, allowing for further analysis of each person's behavior in surveillance videos. Besides the dataset, we propose a novel video captioning approach that can describe individual behavior in detail on a person-level basis, achieving state-of-the-art results. To facilitate further research in this field, we intend to release our dataset and code.

The recent contrastive language-image pre-training (CLIP) model has shown great success in a wide range of image-level tasks, revealing remarkable ability for learning powerful visual representations with rich semantics. An open and worthwhile problem is efficiently adapting such a strong model to the video domain and designing a robust video anomaly detector. In this work, we propose VadCLIP, a new paradigm for weakly supervised video anomaly detection (WSVAD) by leveraging the frozen CLIP model directly without any pre-training and fine-tuning process. Unlike current works that directly feed extracted features into the weakly supervised classifier for frame-level binary classification, VadCLIP makes full use of fine-grained associations between vision and language on the strength of CLIP and involves dual branch. One branch simply utilizes visual features for coarse-grained binary classification, while the other fully leverages the fine-grained language-image alignment. With the benefit of dual branch, VadCLIP achieves both coarse-grained and fine-grained video anomaly detection by transferring pre-trained knowledge from CLIP to WSVAD task. We conduct extensive experiments on two commonly-used benchmarks, demonstrating that VadCLIP achieves the best performance on both coarse-grained and fine-grained WSVAD, surpassing the state-of-the-art methods by a large margin. Specifically, VadCLIP achieves 84.51% AP and 88.02% AUC on XD-Violence and UCF-Crime, respectively. Code and features are released at https://github.com/nwpu-zxr/VadCLIP., Comment: Accept to AAAI2024

Video anomaly detection (VAD) has been paid increasing attention due to its potential applications, its current dominant tasks focus on online detecting anomalies% at the frame level, which can be roughly interpreted as the binary or multiple event classification. However, such a setup that builds relationships between complicated anomalous events and single labels, e.g., ``vandalism'', is superficial, since single labels are deficient to characterize anomalous events. In reality, users tend to search a specific video rather than a series of approximate videos. Therefore, retrieving anomalous events using detailed descriptions is practical and positive but few researches focus on this. In this context, we propose a novel task called Video Anomaly Retrieval (VAR), which aims to pragmatically retrieve relevant anomalous videos by cross-modalities, e.g., language descriptions and synchronous audios. Unlike the current video retrieval where videos are assumed to be temporally well-trimmed with short duration, VAR is devised to retrieve long untrimmed videos which may be partially relevant to the given query. To achieve this, we present two large-scale VAR benchmarks, UCFCrime-AR and XDViolence-AR, constructed on top of prevalent anomaly datasets. Meanwhile, we design a model called Anomaly-Led Alignment Network (ALAN) for VAR. In ALAN, we propose an anomaly-led sampling to focus on key segments in long untrimmed videos. Then, we introduce an efficient pretext task to enhance semantic associations between video-text fine-grained representations. Besides, we leverage two complementary alignments to further match cross-modal contents. Experimental results on two benchmarks reveal the challenges of VAR task and also demonstrate the advantages of our tailored method. Captions are publicly released at https://github.com/Roc-Ng/VAR., Comment: This work has been accepted to the IEEE TIP. Copyright has been transferred

In the next decades, it is necessary to forge new late-universe cosmological probes to precisely constrain the Hubble constant and the equation of state of dark energy simultaneously. In this work, we show that the four typical late-universe cosmological probes, the 21 cm intensity mapping (IM), fast radio burst (FRB), gravitational wave (GW) standard siren, and strong gravitational lensing (SGL), are expected to be forged into useful tools in solving the Hubble tension and exploring dark energy. We propose that the synergy of them is rather important in cosmology. We simulate the 21 cm IM, FRB, GW, and SGL data based on the hypothetical observations of the Hydrogen Intensity and Real-time Analysis eXperiment (HIRAX), the Square Kilometre Array (SKA), the Einstein Telescope (ET), and the Large Synoptic Survey Telescope (LSST), respectively. We find that the four probes show obviously different parameter degeneracy orientations in cosmological constraints, so any combination of them can break the parameter degeneracies and thus significantly improve the constraint precision. The joint 21 cm IM+FRB+GW+SGL data can provide the constraint errors of $\sigma(\Omega_{\rm m})=0.0022$ and $\sigma(H_0)=0.16\ \rm km\ s^{-1}\ Mpc^{-1}$ in the $\Lambda$CDM model, which meet the standard of precision cosmology, i.e., the constraint precision of parameters is better than 1%. In addition, the joint data give $\sigma(w)=0.020$ in the $w$CDM model, and $\sigma(w_0)=0.066$ and $\sigma(w_a)=0.25$ in the $w_0w_a$CDM model, which are all better than the constraints obtained by the CMB+BAO+SN data. We show that the synergy between the four late-universe cosmological probes has magnificent prospects., Comment: 13 pages, 7 figures

One of the most intense air mass transformations on Earth happens when cold air flows from frozen surfaces to much warmer open water in cold-air outbreaks (CAOs), a process captured beautifully in satellite imagery. Despite the ubiquity of the CAO cloud regime over high-latitude oceans, we have a rather poor understanding of its properties, its role in energy and water cycles, and its treatment in weather and climate models. The Cold-air Outbreaks in the Marine Boundary Layer Experiment (COMBLE) was conducted to better understand this regime and its representation in models. COMBLE aimed to examine the relations between surface fluxes, boundary-layer structure, aerosol, cloud and precipitation properties, and mesoscale circulations in marine CAOs. Processes affecting these properties largely fall in a range of scales where boundary-layer processes, convection, and precipitation are tightly coupled, which makes accurate representation of the CAO cloud regime in numerical weather prediction and global climate models most challenging. COMBLE deployed an Atmospheric Radiation Measurement Mobile Facility at a coastal site in northern Scandinavia (69°N), with additional instruments on Bear Island (75°N), from December 2019 to May 2020. CAO conditions were experienced 19% (21%) of the time at the main site (on Bear Island). A comprehensive suite of continuous in situ and remote sensing observations of atmospheric conditions, clouds, precipitation, and aerosol were collected. Because of the clouds’ well-defined origin, their shallow depth, and the broad range of observed temperature and aerosol concentrations, the COMBLE dataset provides a powerful modeling test bed for improving the representation of mixed-phase cloud processes in large-eddy simulations and large-scale models.

In the next decades, it is necessary to forge new late-universe cosmological probes to precisely constrain the Hubble constant and the equation of state of dark energy simultaneously. In this work, we show that the four typical late-universe cosmological probes, the 21 cm intensity mapping (IM), fast radio burst (FRB), gravitational wave (GW) standard siren, and strong gravitational lensing (SGL), are expected to be forged into useful tools in solving the Hubble tension and exploring dark energy. We propose that the synergy of them is rather important in cosmology. We simulate the 21 cm IM, FRB, GW, and SGL data based on the hypothetical observations of the Hydrogen Intensity and Real-time Analysis eXperiment (HIRAX), the Square Kilometre Array (SKA), the Einstein Telescope (ET), and the Large Synoptic Survey Telescope (LSST), respectively. We find that the four probes show obviously different parameter degeneracy orientations in cosmological constraints, so any combination of them can break the parameter degeneracies and thus significantly improve the constraint precision. The joint 21 cm IM+FRB+GW+SGL data can provide the constraint errors of $\sigma(\Omega_{\rm m})=0.0022$ and $\sigma(H_0)=0.16\ \rm km\ s^{-1}\ Mpc^{-1}$ in the $\Lambda$CDM model, which meet the standard of precision cosmology, i.e., the constraint precision of parameters is better than 1%. In addition, the joint data give $\sigma(w)=0.020$ in the $w$CDM model, and $\sigma(w_0)=0.066$ and $\sigma(w_a)=0.25$ in the $w_0w_a$CDM model, which are all better than the constraints obtained by the CMB+BAO+SN data. We show that the synergy between the four late-universe cosmological probes has magnificent prospects., Comment: 13 pages, 7 figures

One of the most intense air mass transformations on Earth happens when cold air flows from frozen surfaces to much warmer open water in cold-air outbreaks (CAOs), a process captured beautifully in satellite imagery. Despite the ubiquity of the CAO cloud regime over high-latitude oceans, we have a rather poor understanding of its properties, its role in energy and water cycles, and its treatment in weather and climate models. The Cold-Air Outbreaks in the Marine Boundary Layer Experiment (COMBLE) was conducted to better understand this regime and its representation in models. COMBLE aimed to examine the relations between surface fluxes, boundary layer structure, aerosol, cloud, and precipitation properties, and mesoscale circulations in marine CAOs. Processes affecting these properties largely fall in a range of scales where boundary layer processes, convection, and precipitation are tightly coupled, which makes accurate representation of the CAO cloud regime in numerical weather prediction and global climate models most challenging. COMBLE deployed an Atmospheric Radiation Measurement Mobile Facility at a coastal site in northern Scandinavia (69°N), with additional instruments on Bear Island (75°N), from December 2019 to May 2020. CAO conditions were experienced 19% (21%) of the time at the main site (on Bear Island). A comprehensive suite of continuous in situ and remote sensing observations of atmospheric conditions, clouds, precipitation, and aerosol were collected. Because of the clouds’ well-defined origin, their shallow depth, and the broad range of observed temperature and aerosol concentrations, the COMBLE dataset provides a powerful modeling testbed for improving the representation of mixed-phase cloud processes in large-eddy simulations and large-scale models.

One of the most intense air mass transformations on Earth happens when cold air flows from frozen surfaces to much warmer open water in cold-air outbreaks (CAOs), a process captured beautifully in satellite imagery. Despite the ubiquity of the CAO cloud regime over high-latitude oceans, we have a rather poor understanding of its properties, its role in energy and water cycles, and its treatment in weather and climate models. The Cold-air Outbreaks in the Marine Boundary Layer Experiment (COMBLE) was conducted to better understand this regime and its representation in models. COMBLE aimed to examine the relations between surface fluxes, boundary-layer structure, aerosol, cloud and precipitation properties, and mesoscale circulations in marine CAOs. Processes affecting these properties largely fall in a range of scales where boundary-layer processes, convection, and precipitation are tightly coupled, which makes accurate representation of the CAO cloud regime in numerical weather prediction and global climate models most challenging. COMBLE deployed an Atmospheric Radiation Measurement Mobile Facility at a coastal site in northern Scandinavia (69°N), with additional instruments on Bear Island (75°N), from December 2019 to May 2020. CAO conditions were experienced 19% (21%) of the time at the main site (on Bear Island). A comprehensive suite of continuous in situ and remote sensing observations of atmospheric conditions, clouds, precipitation, and aerosol were collected. Because of the clouds’ well-defined origin, their shallow depth, and the broad range of observed temperature and aerosol concentrations, the COMBLE dataset provides a powerful modeling test bed for improving the representation of mixed-phase cloud processes in large-eddy simulations and large-scale models.

The 21 cm intensity mapping (IM) technique provides us with an efficient way to observe the cosmic large-scale structure (LSS). From the LSS data, one can use the baryon acoustic oscillation and redshift space distortion to trace the expansion and growth history of the universe, and thus measure the dark energy parameters. In this paper, we make a forecast for cosmological parameter estimation with the synergy of three 21 cm IM experiments. Specifically, we adopt a novel joint survey strategy, FAST ($0

The cosmic curvature $\Omega_{K,0}$, which determines the spatial geometry of the universe, is an important parameter in modern cosmology. Any deviation from $\Omega_{K,0}=0$ would have a profound impact on primordial inflation paradigm and fundamental physics. In this work, we adopt a cosmological model-independent method to test whether $\Omega_{K,0}$ deviates from zero. We use the Gaussian process to reconstruct the reduced Hubble parameter $E(z)$ and the derivative of distance $D'(z)$ from observational data, and then determine $\Omega_{K,0}$ with a null test relation. The cosmic chronometer (CC) Hubble data, baryon acoustic oscillation (BAO) Hubble data, and supernovae Pantheon sample are considered. Our result is consistent with a spatially flat universe within the domain of reconstruction $01$, it tends to favor a closed universe. In this sense, there is still a possibility for a closed universe. We also carry out the null test of the cosmic curvature at $0

Tang M, Gu XZ, Wu PC, Yang XT. J Inflamm Res. 2022;15:2775-2787. The authors have advised that there is an error in affiliation 2 on page 2775, “Yunnan Blood Disease Clinical Medical Research Center” should read “Yunnan Province Clinical Research Center for Hematologic Disease”. The authors apologize for this error. Read the original article

Lingling Lu,1,* Wei Gu,2,* Hang Xie,3 Xu Wang,4 Liming Cao,1 Mingfeng Shan,1 Peng Wu,1 Ye Tian,1 Kai Zhou1 1Department of Infectious Disease, Children’s Hospital of Nanjing Medical University, Nanjing, People’s Republic of China; 2Department of Quality Management, Children’s Hospital of Nanjing Medical University, Nanjing, People’s Republic of China; 3Children’s Hospital of Nanjing Medical University, Nanjing, People’s Republic of China; 4Department of Endocrinology, Children’s Hospital of Nanjing Medical University, Nanjing, People’s Republic of China*These authors contributed equally to this workCorrespondence: Ye Tian; Kai Zhou, Department of Infectious Disease, Children’s Hospital of Nanjing Medical University, Guangzhou Road 72, Nanjing, 210000, People’s Republic of China, Tel +86-21-52862898, Email medtronic@163.com; 18951769720@163.comPurpose: The Chinese government has authorized the emergency use of an inactivated vaccine for COVID-19 in children and adolescents aged 3 to 17 years. This study aimed to investigate parents’ attitudes towards vaccinating their children against COVID-19 and influencing factors.Patients and Methods: Through an online questionnaire survey, we collected self-reported children’s demographic characteristics, physical conditions and parents’ attitudes towards COVID-19 vaccination for children. The parents in the unwilling group received online consultation about the benefits and risks of COVID-19 vaccine and were asked to complete the questionnaire again.Results: A total of 868 participants were recruited from July 2021 to August 2021 in Nanjing, China. Overall, 76.0% of parents were willing to accept vaccination for children. Parents’ willingness increased with children’s age (P=0.018) and height (P=0.034), but decreased if the children fell sick within previous one month (P=0.030). Most of the unwilling parents gave a higher score to the risk of vaccination (53.76 VS 40.18). Uns

Ming Tang,1,* Xue-Zhong Gu,2,* Peng-Chun Wu,3 Xuan-Tao Yang1 1Department of Pathology, the First People’s Hospital of Yunnan Province, the Affiliated Hospital of Kunming University of Science and Technology, Kunming, 650032, People’s Republic of China; 2Department of Hematology, the First People’s Hospital of Yunnan Province, Yunnan Blood Disease Clinical Medical Research Center, the Affiliated Hospital of Kunming University of Science and Technology, Kunming, 650032, People’s Republic of China; 3Department of Pathology, Kunming Kingmed Institute for Clinical Laboratory, Kunming, 650506, People’s Republic of China*These authors contributed equally to this workCorrespondence: Xuan-Tao Yang, Department of Pathology, the First People’s Hospital of Yunnan Province, the Affiliated Hospital of Kunming University of Science and Technology, No. 157 of Jinbi Road, Xishan District, Kunming, 650032, People’s Republic of China, Tel +86 871 63638441, Email yangxt832@126.com Peng-Chun Wu, Department of Pathology, Kunming Kingmed Institute for Clinical Laboratory, K13-K14, Daliyuan Road, Chenggong District, Kunming, 650506, People’s Republic of China, Tel +86 871 68172222, Email wupengchunwpc70@126.comObjective: To investigate the clinicopathological features, and mutations of NRAS, KRAS, BRAF and MAP2K1 genes in extranodal Rosai–Dorfman disease (RDD).Methods: The clinic opathological features of 27 patients with extranodal RDD were retrospectively analyzed, and the NRAS, KRAS, BRAF and MAP2K1 genes mutation were detected by Sanger sequencing.Results: The male to female ratio was 1.7:1. The average age was 46.9 years. There were skin lesions in 12 cases (44.4%) and head and neck lesions in 8 cases (29.6%). Microscopically, those patients with skin RDD had lesions characterized by clear and dark intervals and obvious emperipolesis, while in other parts, the background was more complex. About 2

The U.S. is the second largest contributor to carbon emissions in the world, with its road transport sector being one of the most significant emission sources. However, few studies have been conducted on factors influencing the emissions changes for the U.S. from the perspective of passenger and freight transport. This study aimed to evaluate the carbon emissions from the U.S. road passenger and freight transport sectors, using a Logarithmic Mean Divisia Index approach. Emissions from 2008 to 2017 in the U.S. road transport sector were analysed and key findings include: (1) energy intensity and passenger transport intensity are critical for reducing emissions from road passenger transport, and transport structure change is causing a shift in emissions between different passenger transport modes; and (2) the most effective strategies to reduce carbon emissions in the road freight transport sector are to improve energy intensity and reduce freight transport intensity. Several policy recommendations regarding reducing energy and transport intensity are proposed. The results and policy recommendations are expected to provide useful references for policy makers to form carbon emissions reduction strategies for the road transport sector.

Small data samples are still a critical challenge for spatial predictions. Land use regression (LUR) is a widely used model for spatial predictions with observations at a limited number of locations. Studies have demonstrated that LUR models can overcome the limitation exhibited by other spatial prediction models which usually require greater spatial densities of observations. However, the prediction accuracy and robustness of LUR models still need to be improved due to the linear regression within the LUR model. To improve LUR models, this study develops a land use quantile regression (LUQR) model for more accurate spatial predictions for small data samples. The LUQR is an integration of the LUR and quantile regression, which both have advantages in predictions with a small data set of samples. In this study, the LUQR model is applied in predicting spatial distributions of annual mean PM2.5 concentrations across the Greater Sydney Region, New South Wales, Australia, with observations at 19 valid monitoring stations in 2020. Cross validation shows that the goodness-of-fit can be improved by 25.6–32.1% by LUQR models when compared with LUR, and prediction root mean squared error (RMSE) and mean absolute error (MAE) can be reduced by 10.6–13.4% and 19.4–24.7% by LUQR models, respectively. This study also indicates that LUQR is a more robust model for the spatial prediction with small data samples than LUR. Thus, LUQR has great potentials to be widely applied in spatial issues with a limited number of observations.

One of the most intense air mass transformations on Earth happens when cold air flows from frozen surfaces to much warmer open water in cold-air outbreaks (CAOs), a process captured beautifully in satellite imagery. Despite the ubiquity of the CAO cloud regime over high-latitude oceans, we have a rather poor understanding of its properties, its role in energy and water cycles, and its treatment in weather and climate models. The Cold-Air Outbreaks in the Marine Boundary Layer Experiment (COMBLE) was conducted to better understand this regime and its representation in models. COMBLE aimed to examine the relations between surface fluxes, boundary layer structure, aerosol, cloud, and precipitation properties, and mesoscale circulations in marine CAOs. Processes affecting these properties largely fall in a range of scales where boundary layer processes, convection, and precipitation are tightly coupled, which makes accurate representation of the CAO cloud regime in numerical weather prediction and global climate models most challenging. COMBLE deployed an Atmospheric Radiation Measurement Mobile Facility at a coastal site in northern Scandinavia (69 degrees N), with additional instruments on Bear Island (75 degrees N), from December 2019 to May 2020. CAO conditions were experienced 19% (21%) of the time at the main site (on Bear Island). A comprehensive suite of continuous in situ and remote sensing observations of atmospheric conditions, clouds, precipitation, and aerosol were collected. Because of the clouds' well- defined origin, their shallow depth, and the broad range of observed temperature and aerosol concentrations, the COMBLE dataset provides a powerful modeling testbed for improving the representation of mixed-phase cloud processes in large-eddy simulations and large-scale models.

The integration of data and knowledge from several sources is known as data fusion. When data is only available in a distributed fashion or when different sensors are used to infer a quantity of interest, data fusion becomes essential. In Bayesian settings, a priori information of the unknown quantities is available and, possibly, present among the different distributed estimators. When the local estimates are fused, the prior knowledge used to construct several local posteriors might be overused unless the fusion node accounts for this and corrects it. In this paper, we analyze the effects of shared priors in Bayesian data fusion contexts. Depending on different common fusion rules, our analysis helps to understand the performance behavior as a function of the number of collaborative agents and as a consequence of different types of priors. The analysis is performed by using two divergences which are common in Bayesian inference, and the generality of the results allows to analyze very generic distributions. These theoretical results are corroborated through experiments in a variety of estimation and classification problems, including linear and nonlinear models, and federated learning schemes.

People are likely to engage in collective behaviour online during extreme events, such as the COVID-19 crisis, to express their awareness, actions and concerns. Hong Kong has implemented stringent public health and social measures (PHSMs) to curb COVID-19 epidemic waves since the first COVID-19 case was confirmed on 22 January 2020. People are likely to engage in collective behaviour online during extreme events, such as the COVID-19 crisis, to express their awareness, actions and concerns. Here, we offer a framework to evaluate interactions among individuals emotions, perception, and online behaviours in Hong Kong during the first two waves (February to June 2020) and found a strong correlation between online behaviours of Google search and the real-time reproduction numbers. To validate the model output of risk perception, we conducted 10 rounds of cross-sectional telephone surveys from February 1 through June 20 in 2020 to quantify risk perception levels over time. Compared with the survey results, the estimates of the risk perception of individuals using our network-based mechanistic model capture 80% of the trend of people risk perception (individuals who worried about being infected) during the studied period. We may need to reinvigorate the public by engaging people as part of the solution to live their lives with reduced risk.

LiDAR and camera, as two different sensors, supply geometric (point clouds) and semantic (RGB images) information of 3D scenes. However, it is still challenging for existing methods to fuse data from the two cross sensors, making them complementary for quality 3D object detection (3OD). We propose ImLiDAR, a new 3OD paradigm to narrow the cross-sensor discrepancies by progressively fusing the multi-scale features of camera Images and LiDAR point clouds. ImLiDAR enables to provide the detection head with cross-sensor yet robustly fused features. To achieve this, two core designs exist in ImLiDAR. First, we propose a cross-sensor dynamic message propagation module to combine the best of the multi-scale image and point features. Second, we raise a direct set prediction problem that allows designing an effective set-based detector to tackle the inconsistency of the classification and localization confidences, and the sensitivity of hand-tuned hyperparameters. Besides, the novel set-based detector can be detachable and easily integrated into various detection networks. Comparisons on both the KITTI and SUN-RGBD datasets show clear visual and numerical improvements of our ImLiDAR over twenty-three state-of-the-art 3OD methods., Comment: 12 pages

Post-click conversion rate (CVR) prediction is an essential task for discovering user interests and increasing platform revenues in a range of industrial applications. One of the most challenging problems of this task is the existence of severe selection bias caused by the inherent self-selection behavior of users and the item selection process of systems. Currently, doubly robust (DR) learning approaches achieve the state-of-the-art performance for debiasing CVR prediction. However, in this paper, by theoretically analyzing the bias, variance and generalization bounds of DR methods, we find that existing DR approaches may have poor generalization caused by inaccurate estimation of propensity scores and imputation errors, which often occur in practice. Motivated by such analysis, we propose a generalized learning framework that not only unifies existing DR methods, but also provides a valuable opportunity to develop a series of new debiasing techniques to accommodate different application scenarios. Based on the framework, we propose two new DR methods, namely DR-BIAS and DR-MSE. DR-BIAS directly controls the bias of DR loss, while DR-MSE balances the bias and variance flexibly, which achieves better generalization performance. In addition, we propose a novel tri-level joint learning optimization method for DR-MSE in CVR prediction, and an efficient training algorithm correspondingly. We conduct extensive experiments on both real-world and semi-synthetic datasets, which validate the effectiveness of our proposed methods.

Building a scalable and real-time recommendation system is vital for many businesses driven by time-sensitive customer feedback, such as short-videos ranking or online ads. Despite the ubiquitous adoption of production-scale deep learning frameworks like TensorFlow or PyTorch, these general-purpose frameworks fall short of business demands in recommendation scenarios for various reasons: on one hand, tweaking systems based on static parameters and dense computations for recommendation with dynamic and sparse features is detrimental to model quality; on the other hand, such frameworks are designed with batch-training stage and serving stage completely separated, preventing the model from interacting with customer feedback in real-time. These issues led us to reexamine traditional approaches and explore radically different design choices. In this paper, we present Monolith, a system tailored for online training. Our design has been driven by observations of our application workloads and production environment that reflects a marked departure from other recommendations systems. Our contributions are manifold: first, we crafted a collisionless embedding table with optimizations such as expirable embeddings and frequency filtering to reduce its memory footprint; second, we provide an production-ready online training architecture with high fault-tolerance; finally, we proved that system reliability could be traded-off for real-time learning. Monolith has successfully landed in the BytePlus Recommend product., Comment: ORSUM@ACM RecSys 2022

Large imbalance often exists between the foreground points (i.e., objects) and the background points in outdoor LiDAR point clouds. It hinders cutting-edge detectors from focusing on informative areas to produce accurate 3D object detection results. This paper proposes a novel object detection network by semantical point-voxel feature interaction, dubbed PV-RCNN++. Unlike most of existing methods, PV-RCNN++ explores the semantic information to enhance the quality of object detection. First, a semantic segmentation module is proposed to retain more discriminative foreground keypoints. Such a module will guide our PV-RCNN++ to integrate more object-related point-wise and voxel-wise features in the pivotal areas. Then, to make points and voxels interact efficiently, we utilize voxel query based on Manhattan distance to quickly sample voxel-wise features around keypoints. Such the voxel query will reduce the time complexity from O(N) to O(K), compared to the ball query. Further, to avoid being stuck in learning only local features, an attention-based residual PointNet module is designed to expand the receptive field to adaptively aggregate the neighboring voxel-wise features into keypoints. Extensive experiments on the KITTI dataset show that PV-RCNN++ achieves 81.60$\%$, 40.18$\%$, 68.21$\%$ 3D mAP on Car, Pedestrian, and Cyclist, achieving comparable or even better performance to the state-of-the-arts., Comment: 18 pages, 9 figures