Query Selection Research Articles

Approximate dynamic programming methods in Bayesian preference elicitation

Bayesian preference elicitation is a statistical framework which can aid in complex decision making processes such as engineering design by offering preference-informed recommendations to a decision maker (DM). In this framework, a DM is asked to answer a sequence of queries, called a questionnaire, where in each query they must select their preferred option among two alternatives. A common method to construct the questionnaire is through the D-optimality criterion, which seeks to select queries which lead to more precise estimates of utility model parameters. Past works have proposed and investigated adaptive methods for questionnaire construction using D-optimality as the query selection criterion. All of these adaptive methods have been based on greedy strategies to the best of our knowledge. In this work, we propose and compare various non-greedy methods for adaptive questionnaire construction in the preference elicitation setting. Our methods utilize a combination of enumeration and mixed integer programming techniques to select queries to present to the DM. Through our simulation studies, we find only a marginal improvement in D-efficiency through the use of non-greedy methods, suggesting that greedy methods are sufficient for this problem.

End-to-End Roadway Disease Recognition Based on Transformer Architecture

Road disease is a significant factor causing traffic accidents. Timely detection and recognition of road disease is significant for maintaining road safety and reducing traffic accidents. Therefore, it is urgent to study an automatic pavement disease recognition method with accuracy and real-time. However, existing real-time target detectors generally use a convolutional neural network-based architecture, and these detectors usually require post-processing of NMS, which makes the detectors difficult to optimize and unstable, resulting in a delay in reasoning speed. Therefore, we design a target detection model based on Transformer, which uses MobileNet as the backbone network to simplify the network structure and enables interaction and integration of features through an efficient hybrid encoder, which reduces the computational load and does not reduce the detection accuracy. Iou-aware query selection optimizes the object query vector in the Transformer structure and reduces the number of candidate boxes that the model needs to process. The experimental results show that we have achieved good results on the RDD2022 dataset.

Time-Varying Preference Bandits for Robot Behavior Personalization

Robots are increasingly employed in diverse services, from room cleaning to coffee preparation, necessitating an accurate understanding of user preferences. Traditional preference-based learning allows robots to learn these preferences through iterative queries about desired behaviors. However, these methods typically assume static human preferences. In this paper, we challenge this static assumption by considering the dynamic nature of human preferences and introduce the discounted preference bandit method to manage these changes. This algorithm adapts to evolving human preferences and supports seamless human–robot interaction through effective query selection. Our approach outperforms existing methods in time-varying scenarios across three key performance metrics.

AQSFormer: Adaptive Query Selection Transformer for Real-Time Ship Detection from Visual Images

The Internet of Things (IoT) has emerged as a popular topic in both industrial and academic research. IoT devices are often equipped with rapid response capabilities to ensure seamless communication and interoperability, showing significant potential for IoT-based maritime traffic monitoring and navigation safety tasks. However, this also presents major challenges for maritime surveillance systems. The diversity of IoT devices and variability in collected data are substantial. Visual image ship detection is crucial for maritime tasks, yet it must contend with environmental challenges such as haze and waves that can obscure ship details. To address these challenges, we propose an adaptive query selection transformer (AQSFormer) that utilizes two-dimensional rotational position encoding for absolute positioning and integrates relative positions into the self-attention mechanism to overcome insensitivity to the position. Additionally, the introduced deformable attention module focuses on ship edges, enhancing the feature space resolution. The adaptive query selection module ensures a high recall rate and a high end-to-end processing efficiency. Our method improves the mean average precision to 0.779 and achieves a processing speed of 31.3 frames per second, significantly enhancing both the real-time capabilities and accuracy, proving its effectiveness in ship detection.

On the Pros and Cons of Active Learning for Moral Preference Elicitation

Computational preference elicitation methods are tools used to learn people’s preferences quantitatively in a given context. Recent works on preference elicitation advocate for active learning as an efficient method to iteratively construct queries (framed as comparisons between context-specific cases) that are likely to be most informative about an agent’s underlying preferences. In this work, we argue that the use of active learning for moral preference elicitation relies on certain assumptions about the underlying moral preferences, which can be violated in practice. Specifically, we highlight the following common assumptions (a) preferences are stable over time and not sensitive to the sequence of presented queries, (b) the appropriate hypothesis class is chosen to model moral preferences, and (c) noise in the agent’s responses is limited. While these assumptions can be appropriate for preference elicitation in certain domains, prior research on moral psychology suggests they may not be valid for moral judgments. Through a synthetic simulation of preferences that violate the above assumptions, we observe that active learning can have similar or worse performance than a basic random query selection method in certain settings. Yet, simulation results also demonstrate that active learning can still be viable if the degree of instability or noise is relatively small and when the agent’s preferences can be approximately represented with the hypothesis class used for learning. Our study highlights the nuances associated with effective moral preference elicitation in practice and advocates for the cautious use of active learning as a methodology to learn moral preferences.

SOD‐YOLO: Small Object Detection Network for UAV Aerial Images

With the rapid development of the UAV industry, object detection using UAV has become a research hotspot. However, most current object detection models based on deep learning have large parameter counts and are difficult to deploy on embedded devices with limited memory and computational power. To address this problem, a Small Object Detection network for UAV aerial images SOD‐YOLO based on YOLOv8 is proposed, which can meet the application requirements of resource‐constrained devices while ensuring the detection accuracy. First, cross‐domain fusion attention (CDFA) mechanism is proposed to build the C2f‐Attention module in this paper, which is embedded in the backbone network in order to improve the extraction capability of key object features. Meanwhile, the AIFI_LSPE feature fusion module with improved RT‐DETR and the IoU‐aware query selection mechanism are added to the path aggregation network to improve the accuracy of multi‐scale object detection. In addition, in order to balance the sample size ratio and improve the robustness of the network model, we make a new UAV image dataset named VisDrone2019 Extended Edition (VDEE) using images from the VisDrone2019 and UAVDT public datasets. Finally, Shape‐IoU is used as a loss function to reduce the difference between the object GT frame and the detection frame. Experiments show that SOD‐YOLO has a mAP@0.5 of 42.8% in the VDEE dataset, which is increased by 5.1% over YOLOv8. In the VisDrone2019 dataset mAP@0.5 is 39.2%, an improvement of 5.8% over YOLOv8. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

Real-Time Multimodal 3D Object Detection with Transformers

The accuracy and real-time performance of 3D object detection are key factors limiting its widespread application. While cameras capture detailed color and texture features, they lack depth information compared to LiDAR. Multimodal detection combining both can improve results but incurs significant computational overhead, affecting real-time performance. To address these challenges, this paper presents a real-time multimodal fusion model called Fast Transfusion that combines the benefits of LiDAR and camera sensors and reduces the computational burden of their fusion. Specifically, our Fast Transfusion method uses QConv (Quick Convolution) to replace the convolutional backbones compared to other models. QConv concentrates the convolution operations at the feature map center, where the most information resides, to expedite inference. It also utilizes deformable convolution to better match the actual shapes of detected objects, enhancing accuracy. And the model incorporates EH Decoder (Efficient and Hybrid Decoder) which decouples multiscale fusion into intra-scale interaction and cross-scale fusion, efficiently decoding and integrating features extracted from multimodal data. Furthermore, our proposed semi-dynamic query selection refines the initialization of object queries. On the KITTI 3D object detection dataset, our proposed approach reduced the inference time by 36 ms and improved 3D AP by 1.81% compared to state-of-the-art methods.

Measuring Cybercrime in Calls for Police Service

Conventional police databases contain much information on cybercrime, but extracting it remains a practical challenge. This is because these databases rarely contain labels that could be used to automatically retrieve all cybercrime incidents. In this article, we present a supervised machine learning method for extracting cybercrime incidents in calls for police service datasets. Data from the Korean National Police (2020, 9 months, N = 15 million call logs) is used for the demonstration. We combined methods of keyword query selection, minority oversampling, and majority voting techniques to develop a classifier. Three classification techniques, including Naïve Bayes, linear SVM, and kernel SVM, were tested, and the kernel model was chosen to build the final model (accuracy, 93.4%; F1-score, 92.4). We estimate that cybercrime only represents 4.6% of the cases in the selected dataset (excluding traffic-related incidents), but that it can be prevalent with some crime types. We found, for example, that about three quarters (76%) of all fraud incidents have a cyber dimension. We conclude that the cybercrime classification method proposed in this study can support further research on cybercrime and that it offers considerable advantages over manual or keyword-based approaches.

User-centric design of a 3D search interface for protein-ligand complexes

In this work, we present the frontend of GeoMine and showcase its application, focusing on the new features of its latest version. GeoMine is a search engine for ligand-bound and predicted empty binding sites in the Protein Data Bank. In addition to its basic text-based search functionalities, GeoMine offers a geometric query type for searching binding sites with a specific relative spatial arrangement of chemical features such as heavy atoms and intermolecular interactions. In contrast to a text search that requires simple and easy-to-formulate user input, a 3D input is more complex, and its specification can be challenging for users. GeoMine’s new version aims to address this issue from the graphical user interface perspective by introducing an additional visualization concept and a new query template type. In its latest version, GeoMine extends its query-building capabilities primarily through input formulation in 2D. The 2D editor is fully synchronized with GeoMine’s 3D editor and provides the same functionality. It enables template-free query generation and template-based query selection directly in 2D pose diagrams. In addition, the query generation with the 3D editor now supports predicted empty binding sites for AlphaFold structures as query templates. GeoMine is freely accessible on the ProteinsPlus web server (https://proteins.plus).

Effectiveness of Tree-based Ensembles for Anomaly Discovery: Insights, Batch and Streaming Active Learning

Anomaly detection (AD) task corresponds to identifying the true anomalies among a given set of data instances. AD algorithms score the data instances and produce a ranked list of candidate anomalies. The ranked list of anomalies is then analyzed by a human to discover the true anomalies. Ensemble of tree-based anomaly detectors trained in an unsupervised manner and scoring based on uniform weights for ensembles are shown to work well in practice. However, the manual process of analysis can be laborious for the human analyst when the number of false-positives is very high. Therefore, in many real-world AD applications including computer security and fraud prevention, the anomaly detector must be configurable by the human analyst to minimize the effort on false positives. One important way to configure the detector is by providing true labels (nominal or anomaly) for a few instances. Recent work on active anomaly discovery has shown that greedily querying the top-scoring instance and tuning the weights of ensembles based on label feedback allows us to quickly discover true anomalies. This paper makes four main contributions to improve the state-of-the-art in anomaly discovery using tree-based ensembles. First, we provide an important insight that explains the practical successes of unsupervised tree-based ensembles and active learning based on greedy query selection strategy. We also show empirical results on real-world data to support our insights and theoretical analysis to support active learning. Second, we develop a novel batch active learning algorithm to improve the diversity of discovered anomalies based on a formalism called compact description to describe the discovered anomalies. Third, we develop a novel active learning algorithm to handle streaming data setting. We present a data drift detection algorithm that not only detects the drift robustly, but also allows us to take corrective actions to adapt the anomaly detector in a principled manner. Fourth, we present extensive experiments to evaluate our insights and our tree-based active anomaly discovery algorithms in both batch and streaming data settings. Our results show that active learning allows us to discover significantly more anomalies than state-of-the-art unsupervised baselines, our batch active learning algorithm discovers diverse anomalies, and our algorithms under the streaming-data setup are competitive with the batch setup.

OV-VG: A benchmark for open-vocabulary visual grounding

Open-vocabulary learning has emerged as a cutting-edge research area, particularly in light of the widespread adoption of vision-based foundational models. Its primary objective is to comprehend novel concepts that are not encompassed within a predefined vocabulary. One key facet of this endeavor is Visual Grounding (VG), which entails locating a specific region within an image based on a corresponding language description. While current foundational models excel at various visual language tasks, there is a noticeable absence of models specifically tailored for open-vocabulary visual grounding (OV-VG). This research endeavor introduces novel and challenging OV tasks, namely Open-Vocabulary Visual Grounding (OV-VG) and Open-Vocabulary Phrase Localization (OV-PL). The overarching aim is to establish connections between language descriptions and the localization of novel objects. To facilitate this, we have curated a comprehensive annotated benchmark, encompassing 7272 OV-VG images (comprising 10,000 instances) and 1000 OV-PL images. In our pursuit of addressing these challenges, we delved into various baseline methodologies rooted in existing open-vocabulary object detection (OV-D), VG, and phrase localization (PL) frameworks. Surprisingly, we discovered that state-of-the-art (SOTA) methods often falter in diverse scenarios. Consequently, we developed a novel framework that integrates two critical components: Text-Image Query Selection (TIQS) and Language-Guided Feature Attention (LGFA). These modules are designed to bolster the recognition of novel categories and enhance the alignment between visual and linguistic information. Extensive experiments demonstrate the efficacy of our proposed framework, which consistently attains SOTA performance across the OV-VG task. Additionally, ablation studies provide further evidence of the effectiveness of our innovative models. Codes and datasets will be made publicly available at https://github.com/cv516Buaa/OV-VG.

Response Enhanced Semi-supervised Dialogue Query Generation

Leveraging vast and continually updated knowledge from the Internet has been considered an important ability for a dialogue system. Therefore, the dialogue query generation task is proposed for generating search queries from dialogue histories, which will be submitted to a search engine for retrieving relevant websites on the Internet. In this regard, previous efforts were devoted to collecting conversations with annotated queries and training a query producer (QP) via standard supervised learning. However, these studies still face the challenges of data scarcity and domain adaptation. To address these issues, in this paper, we propose a semi-supervised learning framework -- SemiDQG, to improve model performance with unlabeled conversations. Based on the observation that the search query is typically related to the topic of dialogue response, we train a response-augmented query producer (RA) to provide rich and effective training signals for QP. We first apply a similarity-based query selection strategy to select high-quality RA-generated pseudo queries, which are used to construct pseudo instances for training QP and RA. Then, we adopt the REINFORCE algorithm to further enhance QP, with RA-provided rewards as fine-grained training signals. Experimental results and in-depth analysis of three benchmarks show the effectiveness of our framework in cross-domain and low-resource scenarios. Particularly, SemiDQG significantly surpasses ChatGPT and competitive baselines. Our code is available at \url{https://github.com/DeepLearnXMU/SemiDQG}.

One Seed, Two Birds: A Unified Learned Structure for Exact and Approximate Counting

The modern database has many precise and approximate counting requirements. Nevertheless, a solitary multidimensional index or cardinality estimator is insufficient to cater to the escalating demands across all counting scenarios. Such approaches are constrained either by query selectivity or by the compromise between query accuracy and efficiency. We propose CardIndex, a unified learned structure to solve the above problems. CardIndex serves as a versatile solution that not only functions as a multidimensional learned index for accurate counting but also doubles as an adaptive cardinality estimator, catering to varying counting scenarios with diverse requirements for precision and efficiency. Rigorous experimentation has showcased its superiority. Compared to the state-of-the-art (SOTA) autoregressive data-driven cardinality estimation baselines, our structure achieves training and updating times that are two orders of magnitude faster. Additionally, our CPU-based query estimation latency surpasses GPU-based baselines by two to three times. Notably, the estimation accuracy of low-selectivity queries is up to 314 times better than the current SOTA estimator. In terms of indexing tasks, the construction speed of our structure is two orders of magnitude faster than RSMI and 1.9 times faster than R-tree. Furthermore, it exhibits a point query processing speed that is 3%-17% times faster than RSMI and 1.07 to 2.75 times faster than R-tree and KDB-tree. Range queries under specific loads are 20% times faster than the SOTA indexes.

CCDN-DETR: A Detection Transformer Based on Constrained Contrast Denoising for Multi-Class Synthetic Aperture Radar Object Detection.

The effectiveness of the SAR object detection technique based on Convolutional Neural Networks (CNNs) has been widely proven, and it is increasingly used in the recognition of ship targets. Recently, efforts have been made to integrate transformer structures into SAR detectors to achieve improved target localization. However, existing methods rarely design the transformer itself as a detector, failing to fully leverage the long-range modeling advantages of self-attention. Furthermore, there has been limited research into multi-class SAR target detection. To address these limitations, this study proposes a SAR detector named CCDN-DETR, which builds upon the framework of the detection transformer (DETR). To adapt to the multiscale characteristics of SAR data, cross-scale encoders were introduced to facilitate comprehensive information modeling and fusion across different scales. Simultaneously, we optimized the query selection scheme for the input decoder layers, employing IOU loss to assist in initializing object queries more effectively. Additionally, we introduced constrained contrastive denoising training at the decoder layers to enhance the model's convergence speed and improve the detection of different categories of SAR targets. In the benchmark evaluation on a joint dataset composed of SSDD, HRSID, and SAR-AIRcraft datasets, CCDN-DETR achieves a mean Average Precision (mAP) of 91.9%. Furthermore, it demonstrates significant competitiveness with 83.7% mAP on the multi-class MSAR dataset compared to CNN-based models.

A regret-based query selection strategy for the incremental elicitation of the criteria weights in an SRMP model

A regret-based query selection strategy for the incremental elicitation of the criteria weights in an SRMP model

SVMD-TF-QS: An efficient and novel hybrid methodology for the wind speed prediction

Wind power is gaining significant attention as a renewable and environmentally friendly energy source. However, accurate forecasting of wind speed poses challenges due to its inherent variability and stochastic nature. To address this issue, a novel hybrid model (SVMD-TF-QS) for wind speed prediction (WSP) is proposed in this study. The model combines successive variational mode decomposition (SVMD) with a Transformer (TF) based model that incorporates a novel query selection (QS) mechanism. The SVMD component of the hybrid model offers several improvements, including enhanced mode extraction, adaptive mode determination, robustness against initial values of center frequencies, and improved computational efficiency. By decomposing the wind speed data using SVMD, the transformed data is then fed into the TF-QS model. The proposed approach effectively combines the benefits of the QS mechanism and the Transformer model to accurately predict wind speed while minimizing computational load. This is achieved by introducing a deterministic algorithm within the QS mechanism, which computes a sparse approximation of the attention matrix used in the Transformer model. This further enhances the predictive capabilities of the hybrid model. To evaluate its performance and generalization capability, extensive assessments are conducted using data from two wind farms located in Leicester and Portland. The assessments cover various time periods, including 5 min, 10 min, 15 min, 30 min, 1 h, and 2 h WSP intervals. The results of this study provide robust evidence supporting the effectiveness of the proposed hybrid model in WSP for the diverse wind farms and scenarios.

Active Learning for Handling Missing Data.

Recently, the massive growth of IoT devices and Internet data, which are widely used in many applications, including industry and healthcare, has dramatically increased the amount of free unlabeled data collected. However, this unlabeled data is useless if we want to learn supervised machine learning models. The expensive and time-consuming cost of labeling makes the problem even more challenging. Here, the active learning (AL) technique provides a solution by labeling small but highly informative and representative data, which guarantees a high degree of generalizability over space and improves classification performance with data we have never seen before. The task is more difficult when the active learner has no predefined knowledge, such as initial training data, and when the obtained data is incomplete (i.e., contains missing values). In previous studies, the missing data should first be imputed. Then, the active learner selects from the available unlabeled data, regardless of whether the points were originally observed or imputed. However, selecting inaccurate imputed data points would negatively affect the active learner and prevent it from selecting informative and/or representative points, thus reducing the overall classification performance of the prediction models. This motivated us to introduce a novel query selection strategy that accounts for imputation uncertainty when querying new points. For this purpose, we first introduce a novel multiple imputation method that considers feature importance in selecting the most promising feature groups for missing values estimation. This multiple imputation method provides the ability to quantify the imputation uncertainty of each imputed data point. Furthermore, in each of the two phases of the proposed active learner (exploration and exploitation), imputation uncertainty is taken into account to reduce the probability of selecting points with high imputation uncertainty. We tested the effectiveness of the proposed active learner on different binary and multiclass datasets with different missing rates.

Mitral allografts in tricuspid valve position in modern surgery of acquired heart defects: a systematic review

Introduction: In severe recurrent infectious lesions in the right heart, one of the alternative options for bioprosthetics is to implant mitral allografts (donor heart valves) in the tricuspid valve position. Allografts offer numerous benefits, including a configuration that closely resembles a native valve, adaptability to high systemic pressure, absence of foreign elements that can could act as entry points for infections, and antibacterial treatment without lifelong anticoagulant therapy.Objective: To study the feasibility of using a mitral allograft in the tricuspid position based on the analysis of literature data and our own experience.Methods: Searches were conducted in Medline (PubMed), Google Scholar, and Russian Science Citation Index databases using search queries, keywords, and logical operators. The last session was conducted on December 21, 2022. In the initial selection of aforementioned search queries, a total of 4,446 results were obtained (Google Scholar — 3,080, Medline — 1,050, Russian Science Citation Index — 316). After analyzing titles and abstracts, 4,390 studies were excluded due to the insufficient data. Of the remaining studies, 25 scientific papers were excluded due to their duplication. The final group included 31 studies.Results: Based on a systematic literature review, the article presents aspects of the explantation, preparation, and implantation of the mitral allograft in the tricuspid position at the present stage of biotechnology development. The article highlights the main complications, rare clinical observations, and immediate and long-term postoperative results of mitral allograft implantation in the tricuspid position.Conclusion: This technique presents an interesting solution for patients with severe structural changes in the tricuspid valve, but further research is necessary at this stage of biotechnology development. Received 9 March 2023. Revised 7 September 2023. Accepted 11 September 2023. Funding: The study did not have sponsorship. Conflict of interest: The authors declare no conflict of interest. Contribution of the authors: The authors contributed equally to this article.

Center-Guided Transformer for Panoptic Segmentation

A panoptic segmentation network to predict masks and classes for things and stuff in images is proposed in this work. Recently, panoptic segmentation has been advanced through the combination of the query-based learning and end-to-end learning approaches. Current research focuses on learning queries without distinguishing between thing and stuff classes. We present decoupling query learning to generate effective thing and stuff queries for panoptic segmentation. For this purpose, we adopt different workflows for thing and stuff queries. We design center-guided query selection for thing queries, which focuses on the center regions of individual instances in images, while we set stuff queries as randomly initialized embeddings. Also, we apply a decoupling mask to the self-attention of query features to prevent interactions between things and stuff. In the query selection process, we generate a center heatmap that guides thing query selection. Experimental results demonstrate that the proposed panoptic segmentation network outperforms the state of the art on two panoptic segmentation datasets.

A Novel Hybrid Optimization With Ensemble Constraint Handling Approach for the Optimal Materialized Views

The datawarehouse is extremely challenging to work with, as doing so necessitates a significant investment of both time and space. As a result, it is essential to enable rapid data processing in order to cut down on the amount of time needed to respond to queries that are sent to the warehouse. To effectively solve this problem, one of the significant approaches that should be taken is to take the view of materialization. It is extremely unlikely that all of the views that can be derived from the data will ever be materialized. As a result, view subsets need to be selected intelligently in order to enable rapid data processing for queries coming from a variety of locations. The Materialized view selection problem is addressed by the model that has been proposed. The model is based on the ensemble constraint handling techniques (ECHT). In order to optimize the problem, we must take into account the constraints, which include the self-adaptive penalty, the Epsilon ()-parameter, and the stochastic ranking. For the purpose of making a quicker and more accurate selection of queries from the data warehouse, the proposed model includes the implementation of an innovative algorithm known as the constrained hybrid Ebola with COATI optimization (CHECO) algorithm. For the purpose of computing the best possible fitness, the goals of "processing cost of the query," "response cost," and "maintenance cost" are each defined. The top views are selected by the CHECO algorithm based on whether or not the defined fitness requirements are met. In the final step of the process, the proposed model is compared to the models already in use in order to validate the performance improvement in terms of a variety of performance metrics.