Graph Contrastive Learning (GCL) has proven effective in mitigating data sparsity and enhancing representation learning for recommendation. Yet, most GCL frameworks indiscriminately treat all non-anchor nodes as negatives during contrastive sampling, often leading to the false negative problem where semantically similar nodes are incorrectly repelled. Previous attempts to mitigate this issue rely on predetermined heuristics or local neighborhood mining, which struggle to reliably identify false negatives. More critically, they often overlook authentic user-item interactions for anchoring sample relationships. As a result, this paper presents MACRec, a Multi-View subspace-Alignment framework designed to Calibrate contrastive sampling in GCLbased Recommendation. MACRec comprises three core components: (1) a Multi-View Affinity (MVA) module that captures consistent semantic relations across multiple augmentations via self-expression modeling; (2) a Cross-Subspace Alignment (CSA) mechanism that leverages authentic useritem behavioral interactions to enforce semantic consistency across user and item subspaces; and (3) a Calibrationbased Contrastive Reweighting (CCR) strategy to dynamically down-weight potential false negatives during the contrastive learning process. Extensive experiments on three realworld benchmarks demonstrate that MACRec consistently improves performance across various augmentation backbones, achieving up to 14.55% relative gains.

Graph neural networks (GNNs) excel at modeling graph-structured data but often inherit and amplify biases, leading to substantial efforts in developing fair GNNs. However, most existing approaches assume full access to sensitive attribute information, which is often impractical in real-world scenarios due to privacy concerns or risks of discrimination. To address this limitation, this paper focuses on graph fairness with limited sensitive attribute information, ensuring applicability to real-world contexts where current methods fall short. Specifically, we introduce an innovative fairness optimization strategy, propose a novel framework named FGLISA, and provide a theoretical perspective linking limited sensitive attribute information access to fairness objectives, thus enabling fair graph learning in real-world applications with limited sensitive attribute information. Experiments on diverse real-world datasets and tasks validate the effectiveness of our approach in achieving both fairness and predictive performance.

Hao Wu, Hongru Sun, Wanqing Li, Xinguo Yu, Hao Ming, Xiao Luo, Wenbin Zhang, Jiahong Zhao, Yi Guo, Jie Yang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Zhiyin Yu, Yuchen Mou, Juncheng Yan, Junyu Luo, Chunchun Chen, Xing Wei, Yunhui Liu, Hongru Sun, Yuxing Zhang, Jun Xu, Yatao Bian, Ming Zhang, Wei Ye, Tieke He, Jie Yang, Guanjie Zheng, Zhonghai Wu, Bo Zhang, Lei Bai, Xiao Luo. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Sequential recommendation systems aim to predict users’ next interactions by analyzing historical sequences of their behaviors. However, these systems are vulnerable to adversarial attacks that disrupt input sequences through the injection of noisy or irrelevant interactions. Accordingly, this paper introduces the D ual cascad E d i N formati O n fil TER ing ( DENOTER ) algorithm designed to mitigate the impact of adversarial attacks. Unlike existing recommendation methods, which often overlook the varying importance of different components ( i.e ., items and their associated features) within behavioral sequences, DENOTER employs a two-tier adaptive filtering mechanism. Specifically, at the feature level, a dimension-adjustable, forward-only stochastic model is introduced to dynamically select and preserve essential features, thereby effectively filtering out malicious perturbations. At the item level, a Gumbel-driven sampling strategy is leveraged to selectively retain salient items. Through multi-layer refinement with alternating feature- and item-level filtering, DENOTER progressively reduces the influence of adversarially perturbed items and features within input sequences, enabling robust recommendation without specialized data augmentation or model retraining. Extensive empirical evaluations on five datasets and four encoders demonstrate that DENOTER consistently outperforms existing baselines under diverse adversarial conditions, achieving superior robustness and accuracy even in low-resource and out-of-domain settings.

In this paper, we introduce the IMPortance-awaReness maskIng NeTwork (IMPRINT), a novel approach to enhance the robustness of document retrieval systems against query variations, particularly those containing misspellings. Unlike previous models that treat all query components (words/features) equally, IMPRINT prioritizes the most important components while masking out less relevant ones. Specifically, we propose a Mutual Information-based measure to quantify component importance and integrate it into a dynamic masking mechanism that adjusts the retention probability of each component. Our method is evaluated on a combination of three benchmark datasets and three types of query variations. The experimental results show substantial performance gains compared to state-of-the-art models, achieving an average improvement of 1.2 absolute MRR@10 points in retrieval accuracy.

Graph Contrastive Learning (GCL) emerges as a powerful approach in recommendation systems, leveraging graph structures to learn effective representations. However, existing contrastive sampling strategies often introduce unintended biases, most notably, the misclassification of genuine positive samples as negatives, which undermines representation quality and overall recommendation performance. Accordingly, this paper revisits the conventional contrastive sampling and introduces Negative-Free Sampling for Graph Contrastive Learning (NFS). NFS adopts a two-stage sampling strategy that selectively identifies and utilizes only positive instances during training. By removing reliance on negative samples, it effectively mitigates misclassification bias and improves the semantic alignment between related representations. In addition, a comprehensive theoretical analysis is also provided to establish the robustness of NFS against representation collapse. Experimental results on three benchmarks demonstrate that NFS consistently outperforms or performs state-of-the-art methods, achieving up to a 14.2% relative improvement across evaluated datasets. In addition, a detailed ablation study is also provided to examine how exclusively leveraging positive samples contributes to the efficiency of GCL. The results further demonstrate the plug-and-play nature of the proposed method and its resilience to noisy data.

The Segment Anything Model (SAM) establishes a foundational model for prompt-based image segmentation. SAM interactively segments areas of interest using simple point, box, and/or text prompts. However, applying SAM to non-image segmentation tasks remains an open challenge. For instance, localization of actions in an untrimmed skeleton sequence presents its own diverse set of challenges due to small visual and semantic differences around action boundaries. This work proposes a prompt-based action localization method, referred to as PAL-SAM, by repurposing SAM to locate actions within untrimmed skeleton sequences using frame prompts. To tackle the challenging transitional boundaries between actions, a boundary-aware prompt decoder is introduced to improve localization accuracy. Experiments on the public dataset PKU-MMD confirm that PAL-SAM is able to localize actions with frame prompts and shows promise for extension to other types of prompts.

Coordinating multiple embodied agents in dynamic environments remains a core challenge in artificial intelligence, requiring both perception-driven reasoning and scalable cooperation strategies. While recent works have leveraged large language models (LLMs) for multi-agent planning, a few have begun to explore vision-language models (VLMs) for visual reasoning. However, these VLM-based approaches remain limited in their support for diverse embodiment types. In this work, we introduce VIKI-Bench, the first hierarchical benchmark tailored for embodied multi-agent cooperation, featuring three structured levels: agent activation, task planning, and trajectory perception. VIKI-Bench includes diverse robot embodiments, multi-view visual observations, and structured supervision signals to evaluate reasoning grounded in visual inputs. To demonstrate the utility of VIKI-Bench, we propose VIKI-R, a two-stage framework that fine-tunes a pretrained vision-language model (VLM) using Chain-of-Thought annotated demonstrations, followed by reinforcement learning under multi-level reward signals. Our extensive experiments show that VIKI-R significantly outperforms baselines method across all task levels. Furthermore, we show that reinforcement learning enables the emergence of compositional cooperation patterns among heterogeneous agents. Together, VIKI-Bench and VIKI-R offer a unified testbed and method for advancing multi-agent, visual-driven cooperation in embodied AI systems.

Deep neural networks for skeleton-based human action recognition (HAR) often utilize traditional averaging or maximum temporal pooling to aggregate features by treating all joints and frames equally. However, this approach can excessively aggregate less discriminative or even indiscriminative features into the final feature vectors for recognition. To address this issue, a novel method called asynchronous joint adaptive temporal pooling (AJTP) is introduced in this paper. The method aims to enhance action recognition by identifying a set of informative joints across the temporal dimension and applying a joint-based and asynchronous motion-preservative pooling rather than conventional frame-based pooling. The effectiveness of the proposed AJTP has been empirically validated by integrating it with popular Graph Convolutional Network (GCN) models on three benchmark datasets: NTU RGB+D 120, PKUMMD, and Kinetic400. The results have shown that a GCN model with AJTP substantially improves performance compared to its counterpart GCN model with conventional temporal pooling techniques. The source code is available athttps://github.com/ShanakaRG/AJTP.

Data Augmentation (DA) has become a widely adopted strategy for addressing data scarcity in numerous NLP tasks, especially in scenarios with limited resources or imbalanced classes. However, many existing augmentation techniques rely on randomness or additional resources, presenting challenges in both performance and practical implementation. Furthermore, there is a lack of exploration into what constitutes effective augmentation. In this paper, we systematically evaluate existing DA methods across a comprehensive range of text-classification benchmarks. The empirical analysis highlights that the most significant change resulting from augmentation is observed in the data variance. This observation inspires the proposed approach, termed Mask-for-Data Augmentation (M4DA), which strategically masks tokens from original samples for augmentation. Specifically, M4DA consists of a Variance-Oriented Masker Module (VMM), which ensures an increase in data variances, and a Complexity-Enhanced Selection Module (CSM), designed to select the augmented sample with the highest semantic complexity. The effectiveness of the proposed method is empirically validated across various text-classification benchmarks, including scenarios with limited or full resources and imbalanced classes. Experimental results demonstrate considerable improvements over state-of-the-arts.

Adversarial examples pose significant challenges for Natural Language Processing (NLP) model robustness, often causing notable performance degradation. While various detection methods have been proposed with the aim of differentiating clean and adversarial inputs, they often require fine-tuning with ample data, which is problematic for low-resource scenarios. To alleviate this issue, a Subspace Clustering based Adversarial Detector (termed SCAD) is proposed in this paper, leveraging a union of subspaces to model the clean data distribution. Specifically, SCAD estimates feature distribution across semantic subspaces, assigning unseen examples to the nearest one for effective discrimination. The construction of semantic subspaces does not require many observations and hence ideal for the low-resource setting.

Ad hoc table retrieval refers to the task of performing semantic matching between given queries and candidate tables. In recent years, the approach to addressing this retrieval task has undergone significant shifts, transitioning from utilizing hand-crafted features to leveraging the power of Pre-trained Language Models (PLMs). However, key challenges arise when candidate tables contain shared items, and/or queries may refer to only a subset of table items rather than the entire one. Existing models often struggle to distinguish the most informative items and fail to accurately identify the relevant items required to match with the query.

Fully automatic postural assessment is highly useful, but has been challenging. Conventional methods either require manual assessment by ergonomists or depend on special devices that are intrusive, thus being hardly feasible in daily activities and workplaces. In this work, an attention-based convolutional neural network (CNN) is developed for automatic whole-body postural assessment. The proposed network learns to identify highly relevant regions (or body parts) and extract features automatically. Risk of the posture is estimated from the extracted features accordingly. To evaluate the proposed method, a postural dataset, referred to as pH36M, is created by re-targeting Human3.6M, one of the largest publicly available datasets for pose estimation using the Rapid Entire Body Assessment (REBA) criteria. Experimental results on pH36M demonstrate that proposed method achieves promising performance in comparison to baselines and the average assessment scores are substantially aligned with human assessment with a Kappa value of 0.73.

Sequential recommendation is of great importance for a variety of purposes, such as application engineering, resource optimization, and marketing. Yet, existing sequence-based recommendation models are susceptible to adversarial attacks, which aim to perturb input sequences and mislead trained models, resulting in incorrect predictions. Defense methods are accordingly adopted to enhance model robustness. Nevertheless, these methods encounter challenges, such as error propagation (from the model output to generate adversarial samples), the high system complexity, and the difficulty of maintaining the model generalizability. To bridge this gap, this paper introduces a simple yet effective adversarial defense algorithm, termed Perturbation-Driven Sequential Recommendation (PDSR). In the training process, PDSR leverages a simple perturbation-generation module to create adversarial samples, eliminating the need for gradient estimation, thus streamlining the process. Additionally, it also incorporates a robust encoder designed to increase tolerance towards representation variations by ensuring alignment between original and perturbed representations, thereby boosting model generalizability. Comprehensive experiments are conducted based on a combination of five benchmark datasets, two attack methods, and four sequential recommendation models. When compared to four state-of-the-art defense baselines, PDSR demonstrates notable improvements in defense performance.

Entities and relations extraction is one of the key task to build medical knowledge graph, which is of great significance to the development of medical artificial intelligence. However, overlapping triples are great challenge for biomedical entities and relations extraction. In order to improve the performance of biomedical entities and relations extraction, we propose a joint extraction of entities and relations method based on decomposition and recombination strategy to mine biomedical text. Our method decomposes entities and relations extraction task into three related sub-modules, which are entity tagging module, relation classification module and recombination matching module. Our main contributions are as follows: first, we introduce a decomposition and recombination end-to-end learning framework for joint entities and relations extraction. Second, we propose a bi-directional prediction method to deal with the overlapping triples problem. Finally, we propose the negative samples generation method to alleviate the error accumulation among these modules. The extensive experiments demonstrate that our method can improve the F1 score by 4.36%, 2.13% and 11.72% in ADE, DDI and BB biomedical corpus.

Cyber attacks and intrusions have become the major obstacles to the adoption of the Industrial Internet of Things (IIoT) in critical industries. Imbalanced data distribution is a common problem in IIoT environments that negatively influence machine learning-based intrusion detection systems (IDSs). To address this issue, we introduce EvolCostDeep, a hybrid model of stacked autoencoders (SAE) and convolutional neural networks (CNNs) with a new cost-dependent loss function. The loss function aims to optimize the model’s parameters, where the costs are determined using an evolutionary algorithm. The combination of evolutionary algorithms and deep learning (DL) on Big data hinders the scalability of IIoT IDSs. In this regard, a fog computing-enabled framework, called DeepIDSFog, is designed at the data level, where the master node shares the EvolCostDeep model with worker nodes. In each fog worker node, the EvolCostDeep is parallelized through one task-level and two model-level mechanisms. After aggregating detection outputs from worker nodes to the master, the result is passed to the cloud platform for mitigating attacks. A series of experiments is conducted on the ToN-IoT and UNSW-NB15 data sets to evaluate the performance of EvolCostDeep and DeepIDSFog. The results show that our frameworks can effectively handle both class imbalance problem and scalability of big IIoT traffic data compared with the other models. The averaged values of the EvolCostDeep for recall, precision, and$F1$-score on the data sets are of 93.3%, 97.6%, and 95.2%, respectively, which are higher than the compared methods. Also, the DeepIDSFog provides an average speedup of$38.7\times $over other comparing models.

With the prevalence of smart devices, billions of people are accessing digital resource in their daily life. Online user-behavior modeling, as such, has been actively researched in recent years. However, due to the data uncertainty (sparse-ness and skewness), traditional techniques suffer from certain drawbacks, such as relying on labor-intensive expertise or prior knowledge, lacking of interpretability and transparency, and expensive computational cost. As a step toward bridging the gap, this paper proposes a fuzzy-set based contrastive learning algorithm. The general idea is to design an end-to-end learning framework of optimizing representation from contrastive samples. The proposed algorithm is characterized by three main modules, including data augmentation, fuzzy encoder, and semi-supervised optimization. More precisely, data augmentation is used to produce contrastive (positive and negative) samples based on anchor ones. The fuzzy encoder is introduced to fuzzify (or encode) latent representation of those contrastive samples, while the semi-supervised learning is then implemented to optimize the fuzzy encoder and minimize the training loss simultaneously. The advantage of the proposed algorithm includes no requirement of domain knowledge, preserving the transparency and interpretability of result, and computational effectiveness. Experimental results, based on a real-world app usage dataset, demonstrate that the applicability and flexibility of the proposed algorithm, compared with other state-of-the-art methods.

This article uses a case study of the reception of popular South Korean films in China to investigate aspects of proactive participation in the Chinese social mediasphere. Based on the big data collected from the popular Douban online review platform, the authors provide an in-depth investigation of the characteristics of online opinion leaders and leadership. First, we statistically analyze user-generated content-related features for both leader and non-leader groups. In addition, we employ a Bayesian optimization-based imbalanced learning algorithm to identify characteristics of the minority opinion leaders. The efficiency of the proposed algorithm is evaluated based on 17,786 reviews collected from 35 popular films covering a variety of genres, and the proposed algorithm is shown to perform accurately by identifying opinion leaders, compared to other state-of-the-art approaches. Our study also reveals some major features associated with opinion leaders, such as vocabulary, viewing habits, and friendship networks. Finally, we show how these elements are contributing to new understandings of the rapidly transforming digital domain of Chinese social media and the roles of such platforms.

In this paper, we developed a soft clustering algorithm, named NDFS clustering, based on fuzzy sets defined over a multi-dimensional feature space where each fuzzy set represents a cluster. The NDFS clustering algorithm implements an expectation-maximization (EM) paradigm to solve an optimisation problem with objective function described by the degrees of membership of data points with respect to fuzzy sets and with necessary constraints. Details of the NDFS clustering algorithm are discussed. The NDFS clustering algorithm is then applied to analyse primary school learning performance. The case study demonstrates the flexibility and performance of the NDFS clustering algorithm.

Cross-border E-commerce has grown exponentially in the past decade. To gain global competitivity in product-convergent markets, China's over 200 thousands cross-border E-commerce businesses have focused more on the service and cost of supply chain downstream. Therefore, selecting appropriate cost control strategy has marked impact on them. In this study, we evaluated three strategic cost control measures according to 10 evaluation criteria by using a complex fuzzy set based model, named C-COPRAS. The C-COPRAS model is an extension of the COmplex PRoportion ASessment (COPRAS) method. This model uses complex fuzzy set to tackle uncertainty and temporal features in given evaluation context. We then apply this model to a case study of helping a Chinese E-commerce business to select strategic cost control measure on supply chain downstream.

Modelling technology integration in the teaching and learning environment is a complex, uncertain and dynamic practice. A large amount of student behaviour data has been gathered literately for different processing purposes. Yet, considerable questions are still remaining due to the huge data volume, diversification and uncertainty. In this work, we implement a big-data analytical framework for online behaviour modelling, particularly taking streaming data of students' online activity from their laptop usage as an illustrative example. The proposed framework covers details from accessing streaming records to storing heterogeneous data. Furthermore, the work also demonstrates the use of a TF-IDF based feature generation and fuzzy representation strategy to discover critical patterns via this behaviour data. The accuracy of the modelling work is evaluated using students' score on a national-wide test. Experimental results show that the employed TF-IDF feature is much stabler than other traditional features, thereby achieving a better modelling performance. In summary, the simulation result demonstrates the flexibility and applicability of the proposed framework for processing complex behaviour data, and revealing important patterns for decision making.

Identifying matched local government areas (LGAs) through indicators such as socio-demography, economic and industrial development, and environmental variables has significant effects on public service management. Profiling of LGAs can be a fundamental activity for cooperation between government authorities to assist in the coordination tasks and facilitate decision making in short or long term planning. However, regional profiles contain data with typical temporal and uncertain features which bring challenges to identify matched regions. In this paper, a conceptual model based on complex fuzzy sets has been presented which uses complex fuzzy sets to process the temporal and uncertain features simultaneously. The conceptual model was tested through a local government area clustering problem. Testing results indicate it is flexible and applicable to handle periodic and uncertain data.

Digital technology integration in schools and what this means for teaching and learning plays an significant role in shaping the education environment. There has been a growing body of literature addressing students' perceptions towards technology integration. A large amount of student and teacher self-reported questionnaire or survey data therefore has been collected for different modelling purposes. Yet, considerable questions are still remaining due to this huge-volume, diversified and uncertain survey data. This paper demonstrates the use of fuzzy representation and feature selection to discover unique patterns via survey data. More precisely, fuzzy representation is used to quantify survey response and reform response using linguistic expression. Furthermore, a novel feature selection algorithm is applied to identify important features. This proposed algorithm, based on the sparse representation model, selects features that minimize the residual output error iteratively, thus the resulting features have a direct correspondence to the given problem. The efficiency of the proposed work is evaluated using a state-level student survey. The employed dataset (N = 8528) is used to discover unique patterns among computer efficacy, engagement and school engagement. Experimental results show that the proposed algorithm outperforms traditional approaches.

Transportation infrastructure takes a primary role in urban development planning. To better facilitate or understand the infrastructure status and demands, a huge amount of transportation data such as traffic flow counts has been collected from numerous transportation monitoring systems. Making full use of harvested data samples to discover important patterns has become an increasingly appealing research topic, in which a sophisticated and uncertainty-processing framework is required. In this paper, a big-data processing framework is introduced to analyse the transportation data, particularly taking the classification problem of the parking occupation status as an illustrative example. Three modules are implemented to crawl the raw records, generate high-level features, and apply the machine learning algorithm for classification. In addition, the fuzzification algorithm is also introduced to quantify the key attributes of the data, which helps in removing the data redundancy and inconsistency. The proposed framework then is evaluated using a real-world dataset collected from twelve car parks in a university. Simulation results show that the proposed framework performs well with a convincing classification accuracy.

We address the training problem of the sparse Least Squares Support Vector Machines (SVM) using compressed sensing. The proposed algorithm regards the support vectors as a dictionary and selects the important ones that minimize the residual output error iteratively. A measurement matrix is also introduced to reduce the computational cost. The main advantage is that the proposed algorithm performs model training and support vector selection simultaneously. The performance of the proposed algorithm is tested with several benchmark classification problems in terms of number of selected support vectors and size of the measurement matrix. Simulation results show that the proposed algorithm performs competitively when compared to existing methods.

Over the past several decades, neural networks have evolved into powerful computation systems, which are able to learn complex nonlinear input-output relationship from data. However, the structure optimization problem of neural network is a big challenge for processing huge-volumed, diversified and uncertain data. This paper focuses on this problem and introduces a network pruning algorithm based on sparse representation, termed SRP. The proposed approach starts with a large network, then selects important hidden neurons from the original structure using a forward selection criterion that minimizes the residual output error. Furthermore, the presented algorithm has no constraints on the network type. The efficiency of the proposed approach is evaluated based on several benchmark data sets. We also evaluate the performance of the proposed algorithm on a real-world application of individual travel mode choice. The experimental results have shown that SRP performs favorably compared to alternative approaches.

This paper addresses the problem of Through-the-Wall Radar Imaging (TWRI) using the Multiple-Measurement Vector (MMV) compressive sensing model. TWR image formation is reformulated as a compressed sensing (CS) problem, seeking a sparse representation in the spatial domain. In traditional CS-based through-the-wall radar imaging (TWRI) methods, the measurement matrix is vectorized so that a single measurement vector (SMV) model is applied to generate a sparse solution, which represents a scene comprising point-like targets. For multiple measurement TWRI problems, the SMV model may produce a sub-optimum sparse solution. On the other hand, the proposed MMV model for TWRI generates a more sparse scene by processing all the measurements simultaneously. To evaluate the effectiveness of the proposed method, it is applied to fuse multiple polarization data to form the radar image. Based on simulated data with different number of measurements and noise levels, the proposed MMV-based TWRI method produces better TWR images in terms of image quality and detection accuracy.

The last decade has witnessed a great interest in using evolutionary algorithms, such as genetic algorithms, evolutionary strategies and particle swarm optimization (PSO), for multivariate optimization. This paper presents a hybrid algorithm for searching a complex domain space, by combining the PSO and orthogonal design. In the standard PSO, each particle focuses only on the error propagated back from the best particle, without “communicating” with other particles. In our approach, this limitation of the standard PSO is overcome by using a novel crossover operator based on orthogonal design. Furthermore, instead of the “generating-and-updating” model in the standard PSO, the elitism preservation strategy is applied to determine the possible movements of the candidate particles in the subsequent iterations. Experimental results demonstrate that our algorithm has a better performance compared to existing methods, including five PSO algorithms and three evolutionary algorithms.

Least Squares Support Vector Machine (LS-SVM) has become a fundamental tool in pattern recognition and machine learning. However, the main disadvantage is lack of sparseness of solutions. In this article Compressive Sampling (CS), which addresses the sparse signal representation, is employed to find the support vectors of LS-SVM. The main difference between our work and the existing techniques is that the proposed method can locate the sparse topology while training. In contrast, most of the traditional methods need to train the model before finding the sparse support vectors. An experimental comparison with the standard LS-SVM and existing algorithms is given for function approximation and classification problems. The results show that the proposed method achieves comparable performance with typically a much sparser model.

This paper presents a novel algorithm for the dimensionality reduction which employs compressed sensing (CS) to improve the generalization capability of a classifier, especially for large-scale data. Compared to traditional dimensionality reduction methods, the proposed algorithm makes no use of the problem-dependent parameters, nor does it require additional computation for the eigenvalue decomposition like PCA or LDA. Mathematically, the derived algorithm regards the input features as the dictionary in CS, and selects the features that minimize the residual output error iteratively, thus the resulting features have a direct correspondence to the performance requirements of the given problem. Furthermore, the proposed algorithm can be regarded as a sparse classifier, which selects discriminative features and classifies the training data simultaneously. Experimentally, the CS-based algorithm is tested with a hierarchical visual pattern recognition architecture. The simulation results show that not only does the proposed method utilize only 25% of full features while achieving the test accuracy of the original full architecture, but also its performance is competitive when compared to existing dimensionality reduction methods.

The balance between computational complexity and the architecture bottlenecks the development of neural networks (NNs). An architecture that is too large or too small will influence the performance to a large extent in terms of generalization and computational cost. In the past, saliency analysis has been employed to determine the most suitable structure, however, it is time-consuming and the performance is not robust. In this paper, a family of new algorithms for pruning elements (weighs and hidden neurons) in neural networks is presented based on compressive sampling (CS) theory. The proposed framework makes it possible to locate the significant elements, and hence find a sparse structure, without computing their saliency. Experiment results are presented which demonstrate the effectiveness of the proposed approach.