Search

Abstract The surge in 3D modelling has led to a pronounced research emphasis on the field of 3D shape retrieval. Numerous contemporary approaches have been put forth to tackle this intricate challenge. Nevertheless, effectively addressing the intricacies of cross‐modal 3D shape retrieval remains a formidable undertaking, owing to inherent modality‐based disparities. The authors present an innovative notion—termed “geometric words”—which functions as elemental constituents for representing entities through combinations. To establish the knowledge graph, the authors employ geometric words as nodes, connecting them via shape categories and geometry attributes. Subsequently, a unique graph embedding method for knowledge acquisition is devised. Finally, an effective similarity measure is introduced for retrieval purposes. Importantly, each 3D or 2D entity can anchor its geometric terms within the knowledge graph, thereby serving as a link between cross‐domain data. As a result, the authors’ approach facilitates multiple cross‐domain 3D shape retrieval tasks. The authors evaluate the proposed method's performance on the ModelNet40 and ShapeNetCore55 datasets, encompassing scenarios related to 3D shape retrieval and cross‐domain retrieval. Furthermore, the authors employ the established cross‐modal dataset (MI3DOR) to assess cross‐modal 3D shape retrieval. The resulting experimental outcomes, in conjunction with comparisons against state‐of‐the‐art techniques, clearly highlight the superiority of our approach.

Abstract Counting high‐density objects quickly and accurately is a popular area of research. Crowd counting has significant social and economic value and is a major focus in artificial intelligence. Despite many advancements in this field, many of them are not widely known, especially in terms of research data. The authors proposed a three‐tier standardised dataset taxonomy (TSDT). The Taxonomy divides datasets into small‐scale, large‐scale and hyper‐scale, according to different application scenarios. This theory can help researchers make more efficient use of datasets and improve the performance of AI algorithms in specific fields. Additionally, the authors proposed a new evaluation index for the clarity of the dataset: average pixel occupied by each object (APO). This new evaluation index is more suitable for evaluating the clarity of the dataset in the object counting task than the image resolution. Moreover, the authors classified the crowd counting methods from a data‐driven perspective: multi‐scale networks, single‐column networks, multi‐column networks, multi‐task networks, attention networks and weak‐supervised networks and introduced the classic crowd counting methods of each class. The authors classified the existing 36 datasets according to the theory of three‐tier standardised dataset taxonomy and discussed and evaluated these datasets. The authors evaluated the performance of more than 100 methods in the past five years on different levels of popular datasets. Recently, progress in research on small‐scale datasets has slowed down. There are few new datasets and algorithms on small‐scale datasets. The studies focused on large or hyper‐scale datasets appear to be reaching a saturation point. The combined use of multiple approaches began to be a major research direction. The authors discussed the theoretical and practical challenges of crowd counting from the perspective of data, algorithms and computing resources. The field of crowd counting is moving towards combining multiple methods and requires fresh, targeted datasets. Despite advancements, the field still faces challenges such as handling real‐world scenarios and processing large crowds in real‐time. Researchers are exploring transfer learning to overcome the limitations of small datasets. The development of effective algorithms for crowd counting remains a challenging and important task in computer vision and AI, with many opportunities for future research.

Abstract Underwater object detection (UOD) is crucial for marine economic development, environmental protection, and the planet's sustainable development. The main challenges of this task arise from low‐contrast, small objects, and mimicry of aquatic organisms. The key to addressing these challenges is to focus the model on obtaining more discriminative information. The authors observe that the edges of underwater objects are highly unique and can be distinguished from low‐contrast or mimicry environments based on their edges. Motivated by this observation, an Edge‐guided Representation Learning Network, termed ERL‐Net is proposed, that aims to achieve discriminative representation learning and aggregation under the guidance of edge cues. Firstly, an edge‐guided attention module is introduced to model the explicit boundary information, which generates more discriminative features. Secondly, a hierarchical feature aggregation module is proposed to aggregate the multi‐scale discriminative features by regrouping them into three levels, effectively aggregating global and local information for locating and recognising underwater objects. Finally, a wide and asymmetric receptive field block is proposed to enable features to have a wider receptive field, allowing the model to focus on smaller object information. Comprehensive experiments on three challenging underwater datasets show that our method achieves superior performance on the UOD task.

Abstract Generating co‐speech gestures for interactive digital humans remains challenging because of the indeterministic nature of the problem. The authors observe that gestures generated from speech audio or text by existing neural methods often contain less movement shift than expected, which can be viewed as slow or dull. Thus, a new generative model coupled with memory networks as dynamic dictionaries for speech‐driven gesture generation with improved diversity is proposed. More specifically, the dictionary network dynamically stores connections between text and pose features in a list of key‐value pairs as the memory for the pose generation network to look up; the pose generation network then merges the matching pose features and input audio features for generating the final pose sequences. To make the improvements more accurately measurable, a new objective evaluation metric for gesture diversity that can remove the influence of low‐quality motions is also proposed and tested. Quantitative and qualitative experiments demonstrate that the proposed architecture succeeds in generating gestures with improved diversity.

Abstract Human‐machine cooperative control has become an important area of intelligent driving, where driver intention recognition and dynamic control authority allocation are key factors for improving the performance of cooperative decision‐making and control. In this paper, an online learning method is proposed for human‐machine cooperative control, which introduces a priority control parameter in the reward function to achieve optimal allocation of control authority under different driver intentions and driving safety conditions. Firstly, a two‐layer LSTM‐based sequence prediction algorithm is proposed to recognise the driver's lane change (LC) intention for human‐machine cooperative steering control. Secondly, an online reinforcement learning method is developed for optimising the steering authority to reduce driver workload and improve driving safety. The driver‐in‐the‐loop simulation results show that our method can accurately predict the driver's LC intention in cooperative driving and effectively compensate for the driver's non‐optimal driving actions. The experimental results on a real intelligent vehicle further demonstrate the online optimisation capability of the proposed RL‐based control authority allocation algorithm and its effectiveness in improving driving safety.

Abstract The Equilibrium Optimiser (EO) has been demonstrated to be one of the metaheuristic algorithms that can effectively solve global optimisation problems. Balancing the paradox between exploration and exploitation operations while enhancing the ability to jump out of the local optimum are two key points to be addressed in EO research. To alleviate these limitations, an EO variant named adaptive elite‐guided Equilibrium Optimiser (AEEO) is introduced. Specifically, the adaptive elite‐guided search mechanism enhances the balance between exploration and exploitation. The modified mutualism phase reinforces the information interaction among particles and local optima avoidance. The cooperation of these two mechanisms boosts the overall performance of the basic EO. The AEEO is subjected to competitive experiments with state‐of‐the‐art algorithms and modified algorithms on 23 classical benchmark functions and IEE CEC 2017 function test suite. Experimental results demonstrate that AEEO outperforms several well‐performing EO variants, DE variants, PSO variants, SSA variants, and GWO variants in terms of convergence speed and accuracy. In addition, the AEEO algorithm is used for the edge server (ES) placement problem in mobile edge computing (MEC) environments. The experimental results show that the author’s approach outperforms the representative approaches compared in terms of access latency and deployment cost.

Abstract Image clustering has received significant attention due to the growing importance of image recognition. Researchers have explored Riemannian manifold clustering, which is capable of capturing the non‐linear shapes found in real‐world datasets. However, the complexity of image data poses substantial challenges for modelling and feature extraction. Traditional methods such as covariance matrices and linear subspace have shown promise in image modelling, and they are still in their early stages and suffer from certain limitations. However, these include the uncertainty of representing data using only one Riemannian manifold, limited feature extraction capacity of single kernel functions, and resulting incomplete data representation and redundancy. To overcome these limitations, the authors propose a novel approach called join multiple Riemannian manifold representation and multi‐kernel non‐redundancy for image clustering (MRMNR‐MKC). It combines covariance matrices with linear subspace to represent data and applies multiple kernel functions to map the non‐linear structural data into a reproducing kernel Hilbert space, enabling linear model analysis for image clustering. Additionally, the authors use matrix‐induced regularisation to improve the clustering kernel selection process by reducing redundancy and assigning lower weights to identical kernels. Finally, the authors also conducted numerous experiments to evaluate the performance of our approach, confirming its superiority to state‐of‐the‐art methods on three benchmark datasets.

Abstract Continuous emotion recognition is to predict emotion states through affective information and more focus on the continuous variation of emotion. Fusion of electroencephalography (EEG) and facial expressions videos has been used in this field, while there are with some limitations in current researches, such as hand‐engineered features, simple approaches to integration. Hence, a new continuous emotion recognition model is proposed based on the fusion of EEG and facial expressions videos named residual multimodal Transformer (RMMT). Firstly, the Resnet50 and temporal convolutional network (TCN) are utilised to extract spatiotemporal features from videos, and the TCN is also applied to process the computed EEG frequency power to acquire spatiotemporal features of EEG. Then, a multimodal Transformer is used to fuse the spatiotemporal features from the two modalities. Furthermore, a residual connection is introduced to fuse shallow features with deep features which is verified to be effective for continuous emotion recognition through experiments. Inspired by knowledge distillation, the authors incorporate feature‐level loss into the loss function to further enhance the network performance. Experimental results show that the RMMT reaches a superior performance over other methods for the MAHNOB‐HCI dataset. Ablation studies on the residual connection and loss function in the RMMT demonstrate that both of them is functional.

Abstract Vertical federated learning (VFL) can learn a common machine learning model over vertically partitioned datasets. However, VFL are faced with these thorny problems: (1) both the training and prediction are very vulnerable to stragglers; (2) most VFL methods can only support a specific machine learning model. Suppose that VFL incorporates the features of centralised learning, then the above issues can be alleviated. With that in mind, this paper proposes a new VFL scheme, called FedBoost, which makes private parties upload the compressed partial order relations to the honest but curious server before training and prediction. The server can build a machine learning model and predict samples on the union of coded data. The theoretical analysis indicates that the absence of any private party will not affect the training and prediction as long as a round of communication is achieved. Our scheme can support canonical tree‐based models such as Tree Boosting methods and Random Forests. The experimental results also demonstrate the availability of our scheme.

Abstract Support vector machine (SVM) is a binary classifier widely used in machine learning. However, neglecting the latent data structure in previous SVM can limit the performance of SVM and its extensions. To address this issue, the authors propose a novel SVM with discriminative low‐rank embedding (LRSVM) that finds a discriminative latent low‐rank subspace more suitable for SVM classification. The extension models of LRSVM are introduced by imposing different orthogonality constraints to prevent computational inaccuracies. A detailed derivation of the authors’ iterative algorithms are given that is essentially for solving the SVM on the low‐rank subspace. Additionally, some theorems and properties of the proposed models are presented by the authors. It is worth mentioning that the subproblems of the proposed algorithms are equivalent to the standard or the weighted linear discriminant analysis (LDA) problems. This indicates that the projection subspaces obtained by the authors’ algorithms are more suitable for SVM classification compared to those from the LDA method. The convergence analysis for the authors proposed algorithms are also provided. Furthermore, the authors conduct experiments on various machine learning data sets to evaluate the algorithms. The experiment results show that the authors’ algorithms perform significantly better than other algorithms, which indicates their superior abilities on classification tasks.

Abstract Drug‐drug interaction (DDI) prediction is a crucial issue in molecular biology. Traditional methods of observing drug‐drug interactions through medical experiments require significant resources and labour. The authors present a Medical Knowledge Graph Question Answering (MedKGQA) model, dubbed MedKGQA, that predicts DDI by employing machine reading comprehension (MRC) from closed‐domain literature and constructing a knowledge graph of “drug‐protein” triplets from open‐domain documents. The model vectorises the drug‐protein target attributes in the graph using entity embeddings and establishes directed connections between drug and protein entities based on the metabolic interaction pathways of protein targets in the human body. This aligns multiple external knowledge and applies it to learn the graph neural network. Without bells and whistles, the proposed model achieved a 4.5% improvement in terms of DDI prediction accuracy compared to previous state‐of‐the‐art models on the QAngaroo MedHop dataset. Experimental results demonstrate the efficiency and effectiveness of the model and verify the feasibility of integrating external knowledge in MRC tasks.

Abstract One of the most widely used fuzzy linear programming models is the multi‐objective interval type‐2 fuzzy linear programming (IT2FLP) model, which is of particular importance due to the simultaneous integration of multiple criteria and objectives in a single problem, the fuzzy nature of this type of problems, and thus, its closer similarity to real‐world problems. So far, many studies have been done for the IT2FLP problem with uncertainties of the vagueness type. However, not enough studies have been done regarding the multi‐objective interval type‐2 fuzzy linear programming (MOIT2FLP) problem with uncertainties of the vagueness type. As an innovation, this study investigates the MOIT2FLP problem with vagueness‐type uncertainties, which are represented by membership functions (MFs) in the problem. Depending on the localisation of vagueness in the problem, that is, vagueness in the objective function vector, vagueness in the technological coefficients, vagueness in the resources vector, and any possible combination of them, various problems may arise. Furthermore, to solve problems with MOIT2FLP, first, using the weighted sum method as an efficient and effective method, each of the MOIT2FLP problems is converted into a single‐objective problem. In this research, these types of problems are introduced, their MFs are stated, and different solution methods are suggested. For each of the proposed methods, the authors have provided an example and presented the results in the corresponding tables.

Abstract Pre‐trained multilingual language models (PMLMs) such as mBERT and XLM‐R have shown good cross‐lingual transferability. However, they are not specifically trained to capture cross‐lingual signals concerning sentiment words. This poses a disadvantage for low‐resource languages (LRLs) that are under‐represented in these models. To better fine‐tune these models for sentiment classification in LRLs, a novel intermediate task fine‐tuning (ITFT) technique based on a sentiment lexicon of a high‐resource language (HRL) is introduced. The authors experiment with LRLs Sinhala, Tamil and Bengali for a 3‐class sentiment classification task and show that this method outperforms vanilla fine‐tuning of the PMLM. It also outperforms or is on‐par with basic ITFT that relies on an HRL sentiment classification dataset.

Abstract Image classification algorithms are commonly based on the Independent and Identically Distribution (i.i.d.) assumption, but in practice, the Out‐Of‐Distribution (OOD) problem widely exists, that is, the contexts of images in the model predicting are usually unseen during training. In this case, existing models trained under the i.i.d. assumption are limiting generalisation. Causal inference is an important method to learn the causal associations which are invariant across different environments, thus improving the generalisation ability of the model. However, existing methods usually require partitioning of the environment to learn invariant features, which mostly have imbalance problems due to the lack of constraints. In this paper, we propose a balanced causal learning framework (BCL), starting from how to divide the dataset in a balanced way and the balance of training after the division, which automatically generates fine‐grained balanced data partitions in an unsupervised manner and balances the training difficulty of different classes, thereby enhancing the generalisation ability of models in different environments. Experiments on the OOD datasets NICO and NICO++ demonstrate that BCL achieves stable predictions on OOD data, and we also find that models using BCL focus more accurately on the foreground of images compared with the existing causal inference method, which effectively improves the generalisation ability.

Abstract Recovering 3D human meshes from monocular images is an inherently ill‐posed and challenging task due to depth ambiguity, joint occlusion, and truncation. However, most existing approaches do not model such uncertainties, typically yielding a single reconstruction for one input. In contrast, the ambiguity of the reconstruction is embraced and the problem is considered as an inverse problem for which multiple feasible solutions exist. To address these issues, the authors propose a multi‐hypothesis approach, multi‐hypothesis human mesh recovery (MH‐HMR), to efficiently model the multi‐hypothesis representation and build strong relationships among the hypothetical features. Specifically, the task is decomposed into three stages: (1) generating a reasonable set of initial recovery results (i.e., multiple hypotheses) given a single colour image; (2) modelling intra‐hypothesis refinement to enhance every single‐hypothesis feature; and (3) establishing inter‐hypothesis communication and regressing the final human meshes. Meanwhile, the authors take further advantage of multiple hypotheses and the recovery process to achieve human mesh recovery from multiple uncalibrated views. Compared with state‐of‐the‐art methods, the MH‐HMR approach achieves superior performance and recovers more accurate human meshes on challenging benchmark datasets, such as Human3.6M and 3DPW, while demonstrating the effectiveness across a variety of settings. The code will be publicly available at https://cic.tju.edu.cn/faculty/likun/projects/MH‐HMR.

Abstract Hardware‐based sensing frameworks such as cooperative fuel research engines are conventionally used to monitor research octane number (RON) in the petroleum refining industry. Machine learning techniques are employed to predict the RON of integrated naphtha reforming and isomerisation processes. A dynamic Aspen HYSYS model was used to generate data by introducing artificial uncertainties in the range of ±5% in process conditions, such as temperature, flow rates, etc. The generated data was used to train support vector machines (SVM), Gaussian process regression (GPR), artificial neural networks (ANN), regression trees (RT), and ensemble trees (ET). Hyperparameter tuning was performed to enhance the prediction capabilities of GPR, ANN, SVM, ET and RT models. Performance analysis of the models indicates that GPR, ANN, and SVM with R2 values of 0.99, 0.978, and 0.979 and RMSE values of 0.108, 0.262, and 0.258, respectively performed better than the remaining models and had the prediction capability to capture the RON dependence on predictor variables. ET and RT had an R2 value of 0.94 and 0.89, respectively. The GPR model was used as a surrogate model for fitness function evaluations in two optimisation frameworks based on genetic algorithm and particle swarm method. Optimal parameter values found by the optimisation methodology increased the RON value by 3.52%. The proposed methodology of surrogate‐based optimisation will provide a platform for plant‐level implementation to realise the concept of industry 4.0 in the refinery.

Abstract Although deep convolution neural network (DCNN) has achieved great success in computer vision field, such models are considered to lack interpretability in decision‐making. One of fundamental issues is that its decision mechanism is considered to be a “black‐box” operation. The authors design the binary tree structure convolution (BTSC) module and control the activation level of particular neurons to build the interpretable DCNN model. First, the authors design a BTSC module, in which each parent node generates two independent child layers, and then integrate them into a normal DCNN model. The main advantages of the BTSC are as follows: 1) child nodes of the different parent nodes do not interfere with each other; 2) parent and child nodes can inherit knowledge. Second, considering the activation level of neurons, the authors design an information coding objective to guide neural nodes to learn the particular information coding that is expected. Through the experiments, the authors can verify that: 1) the decision‐making made by both the ResNet and DenseNet models can be explained well based on the "decision information flow path" (known as the decision‐path) formed in the BTSC module; 2) the decision‐path can reasonably interpret the decision reversal mechanism (Robustness mechanism) of the DCNN model; 3) the credibility of decision‐making can be measured by the matching degree between the actual and expected decision‐path.

Abstract Exploiting the hierarchical dependence behind user behaviour is critical for click‐through rate (CRT) prediction in recommender systems. Existing methods apply attention mechanisms to obtain the weights of items; however, the authors argue that deterministic attention mechanisms cannot capture the hierarchical dependence between user behaviours because they treat each user behaviour as an independent individual and cannot accurately express users' flexible and changeable interests. To tackle this issue, the authors introduce the Bayesian attention to the CTR prediction model, which treats attention weights as data‐dependent local random variables and learns their distribution by approximating their posterior distribution. Specifically, the prior knowledge is constructed into the attention weight distribution, and then the posterior inference is utilised to capture the implicit and flexible user intentions. Extensive experiments on public datasets demonstrate that our algorithm outperforms state‐of‐the‐art algorithms. Empirical evidence shows that random attention weights can predict user intentions better than deterministic ones.

Abstract Myocarditis is a significant public health concern because of its potential to cause heart failure and sudden death. The standard invasive diagnostic method, endomyocardial biopsy, is typically reserved for cases with severe complications, limiting its widespread use. Conversely, non‐invasive cardiac magnetic resonance (CMR) imaging presents a promising alternative for detecting and monitoring myocarditis, because of its high signal contrast that reveals myocardial involvement. To assist medical professionals via artificial intelligence, the authors introduce generative adversarial networks ‐ multi discriminator (GAN‐MD), a deep learning model that uses binary classification to diagnose myocarditis from CMR images. Their approach employs a series of convolutional neural networks (CNNs) that extract and combine feature vectors for accurate diagnosis. The authors suggest a novel technique for improving the classification precision of CNNs. Using generative adversarial networks (GANs) to create synthetic images for data augmentation, the authors address challenges such as mode collapse and unstable training. Incorporating a reconstruction loss into the GAN loss function requires the generator to produce images reflecting the discriminator features, thus enhancing the generated images' quality to more accurately replicate authentic data patterns. Moreover, combining this loss function with other regularisation methods, such as gradient penalty, has proven to further improve the performance of diverse GAN models. A significant challenge in myocarditis diagnosis is the imbalance of classification, where one class dominates over the other. To mitigate this, the authors introduce a focal loss‐based training method that effectively trains the model on the minority class samples. The GAN‐MD approach, evaluated on the Z‐Alizadeh Sani myocarditis dataset, achieves superior results (F‐measure 86.2%; geometric mean 91.0%) compared with other deep learning models and traditional machine learning methods.

Abstract Diagnosing individuals with autism spectrum disorder (ASD) accurately faces great challenges in clinical practice, primarily due to the data's high heterogeneity and limited sample size. To tackle this issue, the authors constructed a deep graph convolutional network (GCN) based on variable multi‐graph and multimodal data (VMM‐DGCN) for ASD diagnosis. Firstly, the functional connectivity matrix was constructed to extract primary features. Then, the authors constructed a variable multi‐graph construction strategy to capture the multi‐scale feature representations of each subject by utilising convolutional filters with varying kernel sizes. Furthermore, the authors brought the non‐imaging information into the feature representation at each scale and constructed multiple population graphs based on multimodal data by fully considering the correlation between subjects. After extracting the deeper features of population graphs using the deep GCN(DeepGCN), the authors fused the node features of multiple subgraphs to perform node classification tasks for typical control and ASD patients. The proposed algorithm was evaluated on the Autism Brain Imaging Data Exchange I (ABIDE I) dataset, achieving an accuracy of 91.62% and an area under the curve value of 95.74%. These results demonstrated its outstanding performance compared to other ASD diagnostic algorithms.

Abstract Convolutional neural networks depend on deep network architectures to extract accurate information for image super‐resolution. However, obtained information of these convolutional neural networks cannot completely express predicted high‐quality images for complex scenes. A dynamic network for image super‐resolution (DSRNet) is presented, which contains a residual enhancement block, wide enhancement block, feature refinement block and construction block. The residual enhancement block is composed of a residual enhanced architecture to facilitate hierarchical features for image super‐resolution. To enhance robustness of obtained super‐resolution model for complex scenes, a wide enhancement block achieves a dynamic architecture to learn more robust information to enhance applicability of an obtained super‐resolution model for varying scenes. To prevent interference of components in a wide enhancement block, a refinement block utilises a stacked architecture to accurately learn obtained features. Also, a residual learning operation is embedded in the refinement block to prevent long‐term dependency problem. Finally, a construction block is responsible for reconstructing high‐quality images. Designed heterogeneous architecture can not only facilitate richer structural information, but also be lightweight, which is suitable for mobile digital devices. Experimental results show that our method is more competitive in terms of performance, recovering time of image super‐resolution and complexity. The code of DSRNet can be obtained at https://github.com/hellloxiaotian/DSRNet.

Abstract Transportation systems primarily depend on vehicular flow on roads. Developed countries have shifted towards automated signal control, which manages and updates signal synchronisation automatically. In contrast, traffic in underdeveloped countries is mainly governed by manual traffic light systems. These existing manual systems lead to numerous issues, wasting substantial resources such as time, energy, and fuel, as they cannot make real‐time decisions. In this work, we propose an algorithm to determine traffic signal durations based on real‐time vehicle density, obtained from live closed circuit television camera feeds adjacent to traffic signals. The algorithm automates the traffic light system, making decisions based on vehicle density and employing Faster R‐CNN for vehicle detection. Additionally, we have created a local dataset from live streams of Punjab Safe City cameras in collaboration with the local police authority. The proposed algorithm achieves a class accuracy of 96.6% and a vehicle detection accuracy of 95.7%. Across both day and night modes, our proposed method maintains an average precision, recall, F1 score, and vehicle detection accuracy of 0.94, 0.98, 0.96 and 0.95, respectively. Our proposed work surpasses all evaluation metrics compared to state‐of‐the‐art methodologies.

Abstract Cancer is one of the leading causes of death in the world, with radiotherapy as one of the treatment options. Radiotherapy planning starts with delineating the affected area from healthy organs, called organs at risk (OAR). A new approach to automatic OAR segmentation in the chest cavity in Computed Tomography (CT) images is presented. The proposed approach is based on the modified U‐Net architecture with the ResNet‐34 encoder, which is the baseline adopted in this work. The new two‐branch CS‐SA U‐Net architecture is proposed, which consists of two parallel U‐Net models in which self‐attention blocks with cosine similarity as query‐key similarity function (CS‐SA) blocks are inserted between the encoder and decoder, which enabled the use of consistency regularisation. The proposed solution demonstrates state‐of‐the‐art performance for the problem of OAR segmentation in CT images on the publicly available SegTHOR benchmark dataset in terms of a Dice coefficient (oesophagus—0.8714, heart—0.9516, trachea—0.9286, aorta—0.9510) and Hausdorff distance (oesophagus—0.2541, heart—0.1514, trachea—0.1722, aorta—0.1114) and significantly outperforms the baseline. The current approach is demonstrated to be viable for improving the quality of OAR segmentation for radiotherapy planning.

Abstract Achieving high‐precision extraction of sea islands from high‐resolution satellite remote sensing images is crucial for effective resource development and sustainable management. Unfortunately, achieving such accuracy for sea island extraction presents significant challenges due to the presence of extensive background interference. A more widely applicable noise‐tolerant matched filter (NTMF) scheme is proposed for sea island extraction based on the MF scheme. The NTMF scheme effectively suppresses the background interference, leading to more accurate and robust sea island extraction. To further enhance the accuracy and robustness of the NTMF scheme, a neural dynamics algorithm is supplemented that adds an error integration feedback term to counter noise interference during internal computer operations in practical applications. Several comparative experiments were conducted on various remote sensing images of sea islands under different noisy working conditions to demonstrate the superiority of the proposed neural dynamics algorithm‐assisted NTMF scheme. These experiments confirm the advantages of using the NTMF scheme for sea island extraction with the assistance of neural dynamics algorithm.

Abstract Offline reinforcement learning (RL) aims to learn policies entirely from passively collected datasets, making it a data‐driven decision method. One of the main challenges in offline RL is the distribution shift problem, which causes the algorithm to visit out‐of‐distribution (OOD) samples. The distribution shift can be mitigated by constraining the divergence between the target policy and the behaviour policy. However, this method can overly constrain the target policy and impair the algorithm's performance, as it does not directly distinguish between in‐distribution and OOD samples. In addition, it is difficult to learn and represent multi‐modal behaviour policy when the datasets are collected by several different behaviour policies. To overcome these drawbacks, the authors address the distribution shift problem by implicit policy constraints with energy‐based models (EBMs) rather than explicitly modelling the behaviour policy. The EBM is powerful for representing complex multi‐modal distributions as well as the ability to distinguish in‐distribution samples and OODs. Experimental results show that their method significantly outperforms the explicit policy constraint method and other baselines. In addition, the learnt energy model can be used to indicate OOD visits and alert the possible failure.

Abstract Mahjong, a complex game with hidden information and sparse rewards, poses significant challenges. Existing Mahjong AIs require substantial hardware resources and extensive datasets to enhance AI capabilities. The authors propose a transformer‐based Mahjong AI (Tjong) via hierarchical decision‐making. By utilising self‐attention mechanisms, Tjong effectively captures tile patterns and game dynamics, and it decouples the decision process into two distinct stages: action decision and tile decision. This design reduces decision complexity considerably. Additionally, a fan backward technique is proposed to address the sparse rewards by allocating reversed rewards for actions based on winning hands. Tjong consists of 15M parameters and is trained using approximately 0.5 M data over 7 days of supervised learning on a single server with 2 GPUs. The action decision achieved an accuracy of 94.63%, while the claim decision attained 98.55% and the discard decision reached 81.51%. In a tournament format, Tjong outperformed AIs (CNN, MLP, RNN, ResNet, VIT), achieving scores up to 230% higher than its opponents. Furthermore, after 3 days of reinforcement learning training, it ranked within the top 1% on the leaderboard on the Botzone platform.

Abstract The Internet of Multimedia Things (IoMT) refers to a network of interconnected multimedia devices that communicate with each other over the Internet. Recently, smart healthcare has emerged as a significant application of the IoMT, particularly in the context of knowledge‐based learning systems. Smart healthcare systems leverage knowledge‐based learning to become more context‐aware, adaptable, and auditable while maintaining the ability to learn from historical data. In smart healthcare systems, devices capture images, such as X‐rays, Magnetic Resonance Imaging. The security and integrity of these images are crucial for the databases used in knowledge‐based learning systems to foster structured decision‐making and enhance the learning abilities of AI. Moreover, in knowledge‐driven systems, the storage and transmission of HD medical images exert a burden on the limited bandwidth of the communication channel, leading to data transmission delays. To address the security and latency concerns, this paper presents a lightweight medical image encryption scheme utilising bit‐plane decomposition and chaos theory. The results of the experiment yield entropy, energy, and correlation values of 7.999, 0.0156, and 0.0001, respectively. This validates the effectiveness of the encryption system proposed in this paper, which offers high‐quality encryption, a large key space, key sensitivity, and resistance to statistical attacks.

Abstract Transformer tracking always takes paired template and search images as encoder input and conduct feature extraction and target‐search feature correlation by self and/or cross attention operations, thus the model complexity will grow quadratically with the number of input images. To alleviate the burden of this tracking paradigm and facilitate practical deployment of Transformer‐based trackers, we propose a dual pooling transformer tracking framework, dubbed as DPT, which consists of three components: a simple yet efficient spatiotemporal attention model (SAM), a mutual correlation pooling Transformer (MCPT) and a multiscale aggregation pooling Transformer (MAPT). SAM is designed to gracefully aggregates temporal dynamics and spatial appearance information of multi‐frame templates along space‐time dimensions. MCPT aims to capture multi‐scale pooled and correlated contextual features, which is followed by MAPT that aggregates multi‐scale features into a unified feature representation for tracking prediction. DPT tracker achieves AUC score of 69.5 on LaSOT and precision score of 82.8 on TrackingNet while maintaining a shorter sequence length of attention tokens, fewer parameters and FLOPs compared to existing state‐of‐the‐art (SOTA) Transformer tracking methods. Extensive experiments demonstrate that DPT tracker yields a strong real‐time tracking baseline with a good trade‐off between tracking performance and inference efficiency.

Abstract Detecting brain tumours is complex due to the natural variation in their location, shape, and intensity in images. While having accurate detection and segmentation of brain tumours would be beneficial, current methods still need to solve this problem despite the numerous available approaches. Precise analysis of Magnetic Resonance Imaging (MRI) is crucial for detecting, segmenting, and classifying brain tumours in medical diagnostics. Magnetic Resonance Imaging is a vital component in medical diagnosis, and it requires precise, efficient, careful, efficient, and reliable image analysis techniques. The authors developed a Deep Learning (DL) fusion model to classify brain tumours reliably. Deep Learning models require large amounts of training data to achieve good results, so the researchers utilised data augmentation techniques to increase the dataset size for training models. VGG16, ResNet50, and convolutional deep belief networks networks extracted deep features from MRI images. Softmax was used as the classifier, and the training set was supplemented with intentionally created MRI images of brain tumours in addition to the genuine ones. The features of two DL models were combined in the proposed model to generate a fusion model, which significantly increased classification accuracy. An openly accessible dataset from the internet was used to test the model's performance, and the experimental results showed that the proposed fusion model achieved a classification accuracy of 98.98%. Finally, the results were compared with existing methods, and the proposed model outperformed them significantly.

Abstract Multi‐agent reinforcement learning relies on reward signals to guide the policy networks of individual agents. However, in high‐dimensional continuous spaces, the non‐stationary environment can provide outdated experiences that hinder convergence, resulting in ineffective training performance for multi‐agent systems. To tackle this issue, a novel reinforcement learning scheme, Mutual Information Oriented Deep Skill Chaining (MioDSC), is proposed that generates an optimised cooperative policy by incorporating intrinsic rewards based on mutual information to improve exploration efficiency. These rewards encourage agents to diversify their learning process by engaging in actions that increase the mutual information between their actions and the environment state. In addition, MioDSC can generate cooperative policies using the options framework, allowing agents to learn and reuse complex action sequences and accelerating the convergence speed of multi‐agent learning. MioDSC was evaluated in the multi‐agent particle environment and the StarCraft multi‐agent challenge at varying difficulty levels. The experimental results demonstrate that MioDSC outperforms state‐of‐the‐art methods and is robust across various multi‐agent system tasks with high stability.

Abstract Deep learning has recently become a viable approach for classifying Alzheimer's disease (AD) in medical imaging. However, existing models struggle to efficiently extract features from medical images and may squander additional information resources for illness classification. To address these issues, a deep three‐dimensional convolutional neural network incorporating multi‐task learning and attention mechanisms is proposed. An upgraded primary C3D network is utilised to create rougher low‐level feature maps. It introduces a new convolution block that focuses on the structural aspects of the magnetic resonance imaging image and another block that extracts attention weights unique to certain pixel positions in the feature map and multiplies them with the feature map output. Then, several fully connected layers are used to achieve multi‐task learning, generating three outputs, including the primary classification task. The other two outputs employ backpropagation during training to improve the primary classification job. Experimental findings show that the authors’ proposed method outperforms current approaches for classifying AD, achieving enhanced classification accuracy and other indicators on the Alzheimer's disease Neuroimaging Initiative dataset. The authors demonstrate promise for future disease classification studies.

Abstract Sentiment analysis is a fine‐grained analysis task that aims to identify the sentiment polarity of a specified sentence. Existing methods in Chinese sentiment analysis tasks only consider sentiment features from a single pole and scale and thus cannot fully exploit and utilise sentiment feature information, making their performance less than ideal. To resolve the problem, the authors propose a new method, GP‐FMLNet, that integrates both glyph and phonetic information and design a novel feature matrix learning process for phonetic features with which to model words that have the same pinyin information but different glyph information. Our method solves the problem of misspelling words influencing sentiment polarity prediction results. Specifically, the authors iteratively mine character, glyph, and pinyin features from the input comments sentences. Then, the authors use soft attention and matrix compound modules to model the phonetic features, which empowers their model to keep on zeroing in on the dynamic‐setting words in various positions and to dispense with the impacts of the deceptive‐setting ones. Experiments on six public datasets prove that the proposed model fully utilises the glyph and phonetic information and improves on the performance of existing Chinese sentiment analysis algorithms.

Abstract Attention mechanism has been a successful method for multimodal affective analysis in recent years. Despite the advances, several significant challenges remain in fusing language and its nonverbal context information. One is to generate sparse attention coefficients associated with acoustic and visual modalities, which helps locate critical emotional semantics. The other is fusing complementary cross‐modal representation to construct optimal salient feature combinations of multiple modalities. A Conditional Transformer Fusion Network is proposed to handle these problems. Firstly, the authors equip the transformer module with CNN layers to enhance the detection of subtle signal patterns in nonverbal sequences. Secondly, sentiment words are utilised as context conditions to guide the computation of cross‐modal attention. As a result, the located nonverbal features are not only salient but also complementary to sentiment words directly. Experimental results show that the authors’ method achieves state‐of‐the‐art performance on several multimodal affective analysis datasets.

Abstract Laminectomy is one of the most common posterior spinal operations. Since the lamina is adjacent to important tissues such as nerves, once damaged, it can cause serious complications and even lead to paralysis. In order to prevent the above injuries and complications, ultrasonic bone scalpel and surgical robots have been introduced into spinal laminectomy, and many scholars have studied the recognition method of the bone tissue status. Currently, almost all methods to achieve recognition of bone tissue are based on sensor signals collected by high‐precision sensors installed at the end of surgical robots. However, the previous methods could not accurately identify the state of spinal bone tissue. Innovatively, the identification of bone tissue status was regarded as a time series classification task, and the classification algorithm LSTM‐FCN was used to process fusion signals composed of force and cutting depth signals, thus achieving an accurate classification of the lamina bone tissue status. In addition, it was verified that the accuracy of the proposed method could reach 98.85% in identifying the state of porcine spinal laminectomy. And the maximum penetration distance can be controlled within 0.6 mm, which is safe and can be used in practice.

Abstract A comparative study of two force perception skill learning approaches for robot‐assisted spinal surgery, the impedance model method and the imitation learning (IL) method, is presented. The impedance model method develops separate models for the surgeon and patient, incorporating spring‐damper and bone‐grinding models. Expert surgeons' feature parameters are collected and mapped using support vector regression and image navigation techniques. The imitation learning approach utilises long short‐term memory networks (LSTM) and addresses accurate data labelling challenges with custom models. Experimental results demonstrate skill recognition rates of 63.61%–74.62% for the impedance model approach, relying on manual feature extraction. Conversely, the imitation learning approach achieves a force perception recognition rate of 91.06%, outperforming the impedance model on curved bone surfaces. The findings demonstrate the potential of imitation learning to enhance skill acquisition in robot‐assisted spinal surgery by eliminating the laborious process of manual feature extraction.

Abstract Multi‐object tracking in autonomous driving is a non‐linear problem. To better address the tracking problem, this paper leveraged an unscented Kalman filter to predict the object's state. In the association stage, the Mahalanobis distance was employed as an affinity metric, and a Non‐minimum Suppression method was designed for matching. With the detections fed into the tracker and continuous ‘predicting‐matching’ steps, the states of each object at different time steps were described as their own continuous trajectories. We conducted extensive experiments to evaluate tracking accuracy on three challenging datasets (KITTI, nuScenes and Waymo). The experimental results demonstrated that our method effectively achieved multi‐object tracking with satisfactory accuracy and real‐time efficiency.

Abstract Automatically detecting and locating remote occlusion small objects from the images of complex traffic environments is a valuable and challenging research. Since the boundary box location is not sufficiently accurate and it is difficult to distinguish overlapping and occluded objects, the authors propose a network model with a second‐order term attention mechanism and occlusion loss. First, the backbone network is built on CSPDarkNet53. Then a method is designed for the feature extraction network based on an item‐wise attention mechanism, which uses the filtered weighted feature vector to replace the original residual fusion and adds a second‐order term to reduce the information loss in the process of fusion and accelerate the convergence of the model. Finally, an objected occlusion regression loss function is studied to reduce the problems of missed detections caused by dense objects. Sufficient experimental results demonstrate that the authors’ method achieved state‐of‐the‐art performance without reducing the detection speed. The mAP@.5 of the method is 85.8% on the Foggy_cityscapes dataset and the mAP@.5 of the method is 97.8% on the KITTI dataset.

Abstract As human‐machine interaction (HMI) in healthcare continues to evolve, the issue of trust in HMI in healthcare has been raised and explored. It is critical for the development and safety of healthcare that humans have proper trust in medical machines. Intelligent machines that have applied machine learning (ML) technologies continue to penetrate deeper into the medical environment, which also places higher demands on intelligent healthcare. In order to make machines play a role in HMI in healthcare more effectively and make human‐machine cooperation more harmonious, the authors need to build good human‐machine trust (HMT) in healthcare. This article provides a systematic overview of the prominent research on ML and HMT in healthcare. In addition, this study explores and analyses ML and three important factors that influence HMT in healthcare, and then proposes a HMT model in healthcare. Finally, general trends are summarised and issues to consider addressing in future research on HMT in healthcare are identified.

Abstract Person image generation aims to generate images that maintain the original human appearance in different target poses. Recent works have revealed that the critical element in achieving this task is the alignment of appearance domain and pose domain. Previous alignment methods, such as appearance flow warping, correspondence learning and cross attention, often encounter challenges when it comes to producing fine texture details. These approaches suffer from limitations in accurately estimating appearance flows due to the lack of global receptive field. Alternatively, they can only perform cross‐domain alignment on high‐level feature maps with small spatial dimensions since the computational complexity increases quadratically with larger feature sizes. In this article, the significance of multi‐scale alignment, in both low‐level and high‐level domains, for ensuring reliable cross‐domain alignment of appearance and pose is demonstrated. To this end, a novel and effective method, named Multi‐scale Cross‐domain Alignment (MCA) is proposed. Firstly, MCA adopts global context aggregation transformer to model multi‐scale interaction between pose and appearance inputs, which employs pair‐wise window‐based cross attention. Furthermore, leveraging the integrated global source information for each target position, MCA applies flexible flow prediction head and point correlation to effectively conduct warping and fusing for final transformed person image generation. Our proposed MCA achieves superior performance on two popular datasets than other methods, which verifies the effectiveness of our approach.

Abstract Eye health has become a global health concern and attracted broad attention. Over the years, researchers have proposed many state‐of‐the‐art convolutional neural networks (CNNs) to assist ophthalmologists in diagnosing ocular diseases efficiently and precisely. However, most existing methods were dedicated to constructing sophisticated CNNs, inevitably ignoring the trade‐off between performance and model complexity. To alleviate this paradox, this paper proposes a lightweight yet efficient network architecture, mixed‐decomposed convolutional network (MDNet), to recognise ocular diseases. In MDNet, we introduce a novel mixed‐decomposed depthwise convolution method, which takes advantage of depthwise convolution and depthwise dilated convolution operations to capture low‐resolution and high‐resolution patterns by using fewer computations and fewer parameters. We conduct extensive experiments on the clinical anterior segment optical coherence tomography (AS‐OCT), LAG, University of California San Diego, and CIFAR‐100 datasets. The results show our MDNet achieves a better trade‐off between the performance and model complexity than efficient CNNs including MobileNets and MixNets. Specifically, our MDNet outperforms MobileNets by 2.5% of accuracy by using 22% fewer parameters and 30% fewer computations on the AS‐OCT dataset.

Abstract Non‐destructive detection of wire bonding defects in integrated circuits (IC) is critical for ensuring product quality after packaging. Image‐processing‐based methods do not provide a detailed evaluation of the three‐dimensional defects of the bonding wire. Therefore, a method of 3D reconstruction and pattern recognition of wire defects based on stereo vision, which can achieve non‐destructive detection of bonding wire defects is proposed. The contour features of bonding wires and other electronic components in the depth image is analysed to complete the 3D reconstruction of the bonding wires. Especially to filter the noisy point cloud and obtain an accurate point cloud of the bonding wire surface, a point cloud segmentation method based on spatial surface feature detection (SFD) was proposed. SFD can extract more distinct features from the bonding wire surface during the point cloud segmentation process. Furthermore, in the defect detection process, a directional discretisation descriptor with multiple local normal vectors is designed for defect pattern recognition of bonding wires. The descriptor combines local and global features of wire and can describe the spatial variation trends and structural features of wires. The experimental results show that the method can complete the 3D reconstruction and defect pattern recognition of bonding wires, and the average accuracy of defect recognition is 96.47%, which meets the production requirements of bonding wire defect detection.

Abstract Network embedding aspires to learn a low‐dimensional vector of each node in networks, which can apply to diverse data mining tasks. In real‐life, many networks include rich attributes and temporal information. However, most existing embedding approaches ignore either temporal information or network attributes. A self‐attention based architecture using higher‐order weights and node attributes for both static and temporal attributed network embedding is presented in this article. A random walk sampling algorithm based on higher‐order weights and node attributes to capture network topological features is presented. For static attributed networks, the algorithm incorporates first‐order to k‐order weights, and node attribute similarities into one weighted graph to preserve topological features of networks. For temporal attribute networks, the algorithm incorporates previous snapshots of networks containing first‐order to k‐order weights, and nodes attribute similarities into one weighted graph. In addition, the algorithm utilises a damping factor to ensure that the more recent snapshots allocate a greater weight. Attribute features are then incorporated into topological features. Next, the authors adopt the most advanced architecture, Self‐Attention Networks, to learn node representations. Experimental results on node classification of static attributed networks and link prediction of temporal attributed networks reveal that our proposed approach is competitive against diverse state‐of‐the‐art baseline approaches.

Abstract Cloud computing has drastically changed the delivery and consumption of live streaming content. The designs, challenges, and possible uses of cloud computing for live streaming are studied. A comprehensive overview of the technical and business issues surrounding cloud‐based live streaming is provided, including the benefits of cloud computing, the various live streaming architectures, and the challenges that live streaming service providers face in delivering high‐quality, real‐time services. The different techniques used to improve the performance of video streaming, such as adaptive bit‐rate streaming, multicast distribution, and edge computing are discussed and the necessity of low‐latency and high‐quality video transmission in cloud‐based live streaming is underlined. Issues such as improving user experience and live streaming service performance using cutting‐edge technology, like artificial intelligence and machine learning are discussed. In addition, the legal and regulatory implications of cloud‐based live streaming, including issues with network neutrality, data privacy, and content moderation are addressed. The future of cloud computing for live streaming is examined in the section that follows, and it looks at the most likely new developments in terms of trends and technology. For technology vendors, live streaming service providers, and regulators, the findings have major policy‐relevant implications. Suggestions on how stakeholders should address these concerns and take advantage of the potential presented by this rapidly evolving sector, as well as insights into the key challenges and opportunities associated with cloud‐based live streaming are provided.

Abstract Policy evaluation (PE) is a critical sub‐problem in reinforcement learning, which estimates the value function for a given policy and can be used for policy improvement. However, there still exist some limitations in current PE methods, such as low sample efficiency and local convergence, especially on complex tasks. In this study, a novel PE algorithm called Least‐Squares Truncated Temporal‐Difference learning (LST2D) is proposed. In LST2D, an adaptive truncation mechanism is designed, which effectively takes advantage of the fast convergence property of Least‐Squares Temporal Difference learning and the asymptotic convergence property of Temporal Difference learning (TD). Then, two feature pre‐training methods are utilised to improve the approximation ability of LST2D. Furthermore, an Actor‐Critic algorithm based on LST2D and pre‐trained feature representations (ACLPF) is proposed, where LST2D is integrated into the critic network to improve learning‐prediction efficiency. Comprehensive simulation studies were conducted on four robotic tasks, and the corresponding results illustrate the effectiveness of LST2D. The proposed ACLPF algorithm outperformed DQN, ACER and PPO in terms of sample efficiency and stability, which demonstrated that LST2D can be applied to online learning control problems by incorporating it into the actor‐critic architecture.

Abstract Few‐shot image classification is the task of classifying novel classes using extremely limited labelled samples. To perform classification using the limited samples, one solution is to learn the feature alignment (FA) information between the labelled and unlabelled sample features. Most FA methods use the feature mean as the class prototype and calculate the correlation between prototype and unlabelled features to learn an alignment strategy. However, mean prototypes tend to degenerate informative features because spatial features at the same position may not be equally important for the final classification, leading to inaccurate correlation calculations. Therefore, the authors propose an effective intraclass FA strategy that aggregates semantically similar spatial features from an adaptive reference prototype in low‐dimensional feature space to obtain an informative prototype feature map for precise correlation computation. Moreover, a dual correlation module to learn the hard and soft correlations was developed by the authors. This module combines the correlation information between the prototype and unlabelled features in both the original and learnable feature spaces, aiming to produce a comprehensive cross‐correlation between the prototypes and unlabelled features. Using both FA and cross‐attention modules, our model can maintain informative class features and capture important shared features for classification. Experimental results on three few‐shot classification benchmarks show that the proposed method outperformed related methods and resulted in a 3% performance boost in the 1‐shot setting by inserting the proposed module into the related methods.

Abstract Three‐way concept analysis is an important tool for information processing, and rule acquisition is one of the research hotspots of three‐way concept analysis. However, compared with three‐way concept lattices, three‐way semi‐concept lattices have three‐way operators with weaker constraints, which can generate more concepts. In this article, the problem of rule acquisition for three‐way semi‐concept lattices is discussed in general. The authors construct the finer relation of three‐way semi‐concept lattices, and propose a method of rule acquisition for three‐way semi‐concept lattices. The authors also discuss the set of decision rules and the relationships of decision rules among object‐induced three‐way semi‐concept lattices, object‐induced three‐way concept lattices, classical concept lattices and semi‐concept lattices. Finally, examples are provided to illustrate the validity of our conclusions.

Abstract In traditional secret image sharing schemes, a secret image is shared among shareholders who have the same position. But if the shareholders have two different positions, essential and non‐essential, it is necessary to use essential secret image sharing schemes. In this article, a verifiable essential secret image sharing scheme based on HLRs is proposed. Shareholder's share consists of two parts. The first part is produced by the shareholders, which prevents the fraud of dealers. The second part is a shadow image that is produced by using HLRs and the first part of share. The verification of the first part of the shares is done for the first time by using multilinear and bilinear maps. Also, for verifying shadow images, Bloom Filters are used for the first time. The proposed scheme is more efficient than similar schemes, and for the first part of the shares, has formal security.

Abstract Disturbance observer‐based control method has achieved good results in the car‐following scenario of intelligent and connected vehicle (ICV). However, the gain of conventional extended disturbance observer (EDO)‐based control method is usually set manually rather than adjusted adaptively according to real time traffic conditions, thus declining the car‐following performance. To solve this problem, a car‐following strategy of ICV using EDO adjusted by reinforcement learning is proposed. Different from the conventional method, the gain of proposed strategy can be adjusted by reinforcement learning to improve its estimation accuracy. Since the “equivalent disturbance” can be compensated by EDO to a great extent, the disturbance rejection ability of the car‐following method will be improved significantly. Both Lyapunov approach and numerical simulations are carried out to verify the effectiveness of the proposed method.

Abstract Rough set theory places great importance on approximation accuracy, which is used to gauge how well a rough set model describes a target concept. However, traditional approximation accuracy has limitations since it varies with changes in the target concept and cannot evaluate the overall descriptive ability of a rough set model. To overcome this, two types of average approximation accuracy that objectively assess a rough set model’s ability to approximate all information granules is proposed. The first is the relative average approximation accuracy, which is based on all sets in the universe and has several basic properties. The second is the absolute average approximation accuracy, which is based on undefinable sets and has yielded significant conclusions. We also explore the relationship between these two types of average approximation accuracy. Finally, the average approximation accuracy has practical applications in addressing missing attribute values in incomplete information tables.