Adaptive Network Research Articles

Adaptive Impedance Control of a Human–Robotic System Based on Motion Intention Estimation and Output Constraints

The rehabilitation exoskeleton represents a typical human–robot system featuring complex nonlinear dynamics. This paper is devoted to proposing an adaptive impedance control strategy for a rehabilitation exoskelton. The patient’s motion intention is estimated online by the neural network (NN) to cope with the intervention of the patient’s subjective motor awareness in the late stage of rehabilitation training. Due to the differences in impedance parameters for training tasks in individual patients and periods, the least square method was used to learn the impedance parameters of the patient. Considering the uncertainties of the exoskeleton and the safety of rehabilitation training, an adaptive neural network impedance controller with output constraints was designed. The NN was applied to approximate the unknown dynamics and the barrier Lyapunov function was applied to prevent the system from violating the output rules. The feasibility and effectiveness of the proposed strategy were verified by simulation.

Adaptive graph diffusion networks: compact and expressive GNNs with large receptive fields

Graph neural networks (GNNs) are widely used in graph-based tasks, but deep GNNs often suffer from oversmoothing. Existing effective deep GNNs have various shortcomings including redundant complexity, oversimplified architecture, or predefined parameters. To address these issues, we propose adaptive graph diffusion networks (AGDNs), a class of compact and expressive GNNs that can effectively leverage deep neighborhood information. We introduce hopwise attention and hopwise convolution with positional embeddings for learning nodewise and channelwise hop weights, respectively, which overcomes oversmoothing and ensures a powerful ability to learn arbitrary filters in the spectral domain. Our experiments demonstrate that AGDNs can effectively learn various filters on images and exhibit superior performance on diverse and challenging open graph benchmark datasets for node classification and link prediction tasks while maintaining moderate complexity and fast running time.

Efficient Estimation for Longitudinal Networks via Adaptive Merging

Longitudinal networks consist of sequences of temporal edges among multiple nodes, where the temporal edges are observed in real-time. They have become ubiquitous with the rise of online social platforms and e-commerce, but largely under-investigated in the literature. In this paper, we propose an efficient estimation framework for longitudinal networks, leveraging strengths of adaptive network merging, tensor decomposition, and point processes. It merges neighboring sparse networks so as to enlarge the number of observed edges and reduce estimation variance, whereas the estimation bias introduced by network merging is controlled by exploiting local temporal structures for adaptive network neighborhood. A projected gradient descent algorithm is proposed to facilitate estimation, where the upper bound of the estimation error in each iteration is established. A thorough analysis is conducted to quantify the asymptotic behavior of the proposed method, which shows that it can significantly reduce the estimation error and also provides a guideline for network merging under various scenarios. We further demonstrate the advantage of the proposed method through extensive numerical experiments on synthetic datasets and a militarized interstate dispute dataset.

BCS: A neural distinguisher method based on differential propagation uncertainty of nonlinear components and network adaptability

Abstract The neural distinguisher (ND) is the combined product of differential cryptanalysis and deep learning. Its emergence has greatly promoted the development of differential cryptanalysis. Current approaches to improving the performance of NDs focus on data input formats and training frameworks. However, many researchers independently focused on enhancing the data input format or training framework, neglecting their adaptability to each other. Additionally, little research has focused on improving the data input format based on its correlation with the components of the cipher. This paper proposes an ND called the Block Cipher with S-box (BCS) model to address these issues. The model uses new data input formats, Add S-box Multiple Ciphertext Pairs (ASMCP) and Add S-box Multiple Output Difference (ASMOD), along with an improved U-Net network. These two formats are developed based on the decapsulated encryption process method and incorporate the input and output features of the S-box component to increase the available features. The improved U-Net network incorporates long-range skip-connections, better suited for new data input formats. Comparative experiments demonstrate that the BCS model achieves higher distinguishing accuracy and reduces the model size. Finally, the BCS model is applied to key recovery attacks on 8-round PRESENT with a success rate of 98%.

A hybrid CMOS-memristor based adaptable Bidirectional Associative Memory neural network for pattern recognition applications

Abstract This research presents a circuit-level hybrid CMOS memristor architecture for constructing Bidirectional Associative Memory (BAM). Initially, a synaptic circuit structure was built by employing a voltage threshold memristor in a crossbar architecture. This synaptic structure is adaptable and flexible for generating a wide range of synaptic weights. It is then deployed in the BAM network to perform an associative function. To aid in better name recall, this BAM network has been trained to associate Greek and mathematical symbols with their first letters in English, and vice versa. The designed circuit was validated using MATLAB and the EDA (Electronic Design Automation) Tool: Cadence Virtuoso. The addition of noise further evaluates the performance of the BAM network. When tested with noise levels of 10%, 20%, and 30%, the input patterns were retrieved at 100% in both directions. Furthermore, the proposed synaptic circuit is validated for variations in RON, ROFF and its performance is compared with other memristor models. It is also found that the average power consumption of the proposed synaptic circuit is 1.22mW. These results, which were experimentally confirmed, demonstrate the precision and noise isolation of the proposed BAM design. With appropriate tuning of memristor, the synaptic weights can be mapped easily with the memristor conductance value. This circuit can be effectively used in the field of image processing, neural network, and neuromorphic computation, which helps to associate and restore original or damaged binary images, showing strong robustness and accuracy.

Low-dimensional model for adaptive networks of spiking neurons

Low-dimensional model for adaptive networks of spiking neurons

MEGA-GO: Functions prediction of diverse protein sequence length using Multi-scalE Graph Adaptive neural network.

The increasing accessibility of large-scale protein sequences through advanced sequencing technologies has necessitated the development of efficient and accurate methods for predicting protein function. Computational prediction models have emerged as a promising solution to expedite the annotation process. However, despite making significant progress in protein research, graph neural networks face challenges in capturing long-range structural correlations and identifying critical residues in protein graphs. Furthermore, existing models have limitations in effectively predicting the function of newly sequenced proteins that are not included in protein interaction networks. This highlights the need for novel approaches integrating protein structure and sequence data. We introduce MEGA-GO, highlighting the capability of capturing diverse protein sequence length features from multiple scales. The unique graph adaptive neural network architecture of MEGA-GO enables a more nuanced extraction of graph structure features, effectively capturing intricate relationships within biological data. Experimental results demonstrate that MEGA-GO outperforms mainstream protein function prediction models in the accuracy of Gene Ontology (GO) term classification, yielding 33.4%, 68.9%, and 44.6% of Area Under the Precision-Recall Curve (AUPR) on Biological Process (BP), Molecular Function (MF), and Cellular Component (CC) domains respectively. The rest of the experimental results reveal that our model consistently surpasses the state-of-the-art methods. The source code and implementation of MEGA-GO are available at https://github.com/Cheliosoops/MEGA-GO. The supplementary file can be found at Supplementary.

Adaptive neural network tracking control for robotic manipulator with input dead zone and function constraints on states

Adaptive neural network tracking control for robotic manipulator with input dead zone and function constraints on states

Enhancing concealed object detection in active THz security images with adaptation-YOLO

The terahertz (THz) security scanner offers advantages such as non-contact inspection and the ability to detect various types of dangerous goods, playing an important role in preventing terrorist attacks. We aim to accurately and quickly detect concealed objects in THz security images. However, current object detection algorithms face many challenges when applied to THz images. The main reasons for the detection difficulty are that the concealed objects are small, the image resolution is low, and there is back-ground noise. Many methods often ignore the contextual dependency of the objects, hindering the effective capture of the object’s features. To address this task, this paper first proposes an adaptive context-aware attention network (ACAN), which models global contextual association features in both spatial and channel dimensions. By dynamically combining local features and their global relationships, contextual association information can be obtained from the input features, and enhanced attention features can be achieved through feature fusion to enable precise detection of concealed objects. Secondly, we improved the adaptive convolution and developed the dynamic adaptive convolution block (DACB). DACB can adaptively adjust convolution filter parameters and allocate the filters to the corresponding spatial regions, then filter the feature maps to suppress interference information. Finally, we integrated these two components to YOLOv8, resulting in Adaptation-YOLO. Through wide-ranging experiments on the active THz image dataset, the results demonstrate that the suggested method effectively improves the accuracy and efficiency of object detectors.

Epoxy vitrimers: from essence to utility

Abstract Concerns about the effect on the environment and non-renewable nature of plastics have sparked a substantial field of study towards the creation of recyclable polymers. Vitrimers are a potential class of reusable polymers that have recently attracted a lot of interest. Like conventional thermosets in strength, durability, and chemical resistance, these materials offer the added benefit of being recyclable at the end of their useful life. Their chemical structure, which includes dynamic covalent crosslinks to provide stability while enabling reprocessing, is credited with this special characteristic. We lay out an overview of recent developments and their applications in epoxy based vitrimeric materials by using the different types of covalent adaptable networks (CANs) – single, dual and triple in this paper with a lot of attention on design tactics that make it easier to create circular materials of the future. Covalent Adaptable Networks (CAN), a novel polymer family that can bridge the gap between thermosets and thermoplastics, emerged in the recent years and uses dynamic covalent chemistry to crosslinked polymer networks. The field was enhanced in 2011 by Leibler and colleagues when they introduced the notion of vitrimers, which are crosslinked polymers that retain the integrity of their network even after heating and allow the covalent connections to be reallocated within them by associative exchange reactions. This review also demonstrates how the vitrimer community is paying attention to the need for sustainable material development by demonstrating the use of biobased building blocks in the synthesis of novel and high-performing vitrimers. Having outlined the primary characteristics of vitrimers, commercialization and development of vitrimers for different applications is emphasized to portray their benefits for self-healing, malleability, orthogonal processability, and various shape memories along with sustainable solutions to synthetic materials.

C MAL: cascaded network-guided class-balanced multi-prototype auxiliary learning for source-free domain adaptive medical image segmentation.

Source-free domain adaptation (SFDA) has become crucial in medical image analysis, enabling the adaptation of source models across diverse datasets without labeled target domain images. Self-training, a popular SFDA approach, iteratively refines self-generated pseudo-labels using unlabeled target domain data to adapt a pre-trained model from the source domain. However, it often faces model instability due to incorrect pseudo-label accumulation and foreground-background class imbalance. This paper presents a pioneering SFDA framework, named cascaded network-guided class-balanced multi-prototype auxiliary learning (C MAL), to enhance model stability. Firstly, we introduce the cascaded translation-segmentation network (CTS-Net), which employs iterative learning between translation and segmentation networks to generate accurate pseudo-labels. The CTS-Net employs a translation network to synthesize target-like images from unreliable predictions of the initial target domain images. The synthesized results refine segmentation network training, ensuring semantic alignment and minimizing visual disparities. Subsequently, reliable pseudo-labels guide the class-balanced multi-prototype auxiliary learning network (CMAL-Net) for effective model adaptation. CMAL-Net incorporates a new multi-prototype auxiliary learning strategy with a memory network to complement source domain data. We propose a class-balanced calibration loss and multi-prototype-guided symmetry cross-entropy loss to tackle class imbalance issue and enhance model adaptability to the target domain. Extensive experiments on four benchmark fundus image datasets validate the superiority of C MAL over state-of-the-art methods, especially in scenarios with significant domain shifts. Our code is available at https://github.com/yxk-art/C2MAL .

Spatiotemporal Interaction Based Dynamic Adversarial Adaptive Graph Neural Network for Air-Quality Prediction

Air quality issues have become a major environmental concern, with severe air pollution significantly reducing air quality and posing threats to human health. Accurate air quality prediction is crucial for preventing individuals from suffering the detrimental effects of severe air pollution. Recently, deep learning methods based on spatiotemporal graph neural networks (GNNs) have made considerable progress in modeling the temporal and spatial dependencies within air quality data by integrating GNNs with sequential models. Unfortunately, previous work often treats temporal and spatial dependencies as independent components, neglecting the intricate interactions between them. This oversight prevents the models from fully exploiting the complex spatiotemporal dependencies in the data, adversely affecting their predictive performance. To address these issues, we propose a general spatiotemporal interaction framework for air quality prediction. This framework models the bidirectional interactions between temporal and spatial dependencies in a data-driven manner. Furthermore, we designed a spatiotemporal feature extraction module and a dynamic adversarial adaptive graph learning module based on this framework. We introduce the Spatial-Temporal Interaction based Dynamic Adversarial Adaptive Graph Neural Network, capable of capturing the complex interactions between spatiotemporal dependencies and learning the dynamic spatial topology among sites by incorporating the competitive optimization concept of generative adversarial networks. Extensive experiments on two real-world datasets demonstrate the effectiveness of the proposed method, outperforming existing baseline models.

Stereochemistry-Tuned Hydrogen-Bonding Synergistic Covalent Adaptable Networks: Towards Recycled Elastomers with Excellent Creep-Resistant Performance.

Covalent adaptable networks (CANs) offer innovative solutions for the reprocessing and recycling of thermoset polymers. However, achieving a balance between easy reprocessing and creep resistance remains a challenge. This study focuses on designing and synthesizing polyurethane (PU) materials with tailored properties by manipulating the stereochemistry of diamine chain extenders. By employing cis- and trans-configurations of diamine extenders, we developed a series of PU materials with varying mechanical properties and creep resistance. The trans-configured materials (R,R-DAC-PU or S,S-DAC-PU) exhibited superior creep resistance and mechanical strength due to dense hydrogen bonding networks. The cis-configured materials (Cis-DAC-PU) exhibited enhanced processability and elasticity. Under 0.1 MPa stress, R,R-DAC-PU showed a mere 3.5% strain change at 170 °C over 60 minutes, highlighting its superior creep resistance. Both configurations can be recycled via urea bond exchange reactions using hot pressing or solvothermal methods. Density Functional Theory (DFT) calculations indicate that both the (R,R-DCA-UB-U)₂ and (S,S-DCA-UB-U)₂ segments form six hydrogen bonds with shorter bond lengths, leading to stronger hydrogen-bonding interactions. Conversely, the (Cis-DCA-UB-U)₂ segment forms four hydrogen bonds with longer bond lengths, resulting in weaker interactions. This work highlights the critical role of stereochemistry in designing high-performance, recyclable polymer materials with tailored properties.

Stereochemistry‐Tuned Hydrogen‐Bonding Synergistic Covalent Adaptable Networks: Towards Recycled Elastomers with Excellent Creep‐Resistant Performance

Stereochemistry‐Tuned Hydrogen‐Bonding Synergistic Covalent Adaptable Networks: Towards Recycled Elastomers with Excellent Creep‐Resistant Performance

Normalization-Guided and Gradient-Weighted Unsupervised Domain Adaptation Network for Transfer Diagnosis of Rolling Bearing Faults Under Class Imbalance

Normalization-Guided and Gradient-Weighted Unsupervised Domain Adaptation Network for Transfer Diagnosis of Rolling Bearing Faults Under Class Imbalance

CPJN: News recommendation with a content and popularity joint network.

Users may click on a news because they are interested in its content or because the news contains important information and is very popular. Modeling these two aspects is crucial for accurate news recommendation. Most existing studies focused on capturing users' preferences towards news content, and thus they are limited in investigating in depth users' preferences towards news popularity and independently capturing user content and popularity preferences. In this article, we further improve recommendation performance by proposing a news recommendation with content and popularity joint network (CPJN) model. The CPJN contains a content-based network, a popularity-based network, and an adaptive combination network. The content-based network generates a users' preference feature towards news content by eliminating popularity bias in important information extracted from user side information (such as city and age) and uses the information with the eliminated popularity bias to enhance users' preference representation towards news content. The popularity-based network generates a user preference feature towards news popularity by eliminating content bias that is enhanced through news side information (such as category and author). Furthermore, since users exhibit differing degrees of sensitivity towards news popularity, we propose an adaptive combination network to integrate these two preferences for recommendation. Extensive experiments on two real-world datasets demonstrate the effectiveness of CPJN. Compared to the state-of-the-art baseline, CPJN achieved average improvements of 1.493 % in accuracy rate and 1.502 % in recall rate.

Double-level discriminative domain adaptation network for cross-domain fault diagnosis

Double-level discriminative domain adaptation network for cross-domain fault diagnosis

Predictive genetic circuit design for phenotype reprogramming in plants

Plants, with intricate molecular networks for environmental adaptation, offer groundbreaking potential for reprogramming with predictive genetic circuits. However, realizing this goal is challenging due to the long cultivation cycle of plants, as well as the lack of reproducible, quantitative methods and well-characterized genetic parts. Here, we establish a rapid (~10 days), quantitative, and predictive framework in plants. A group of orthogonal sensors, modular synthetic promoters, and NOT gates are constructed and quantitatively characterized. A predictive model is developed to predict the designed circuits’ behavior accurately. Our versatile and robust framework, validated by constructing 21 two-input circuits with high prediction accuracy (R2 = 0.81), enables multi-state phenotype control in both Arabidopsis thaliana and Nicotiana benthamiana in response to chemical inducers. Our study achieves predictable design and application of synthetic circuits in plants, offering valuable tools for the rapid engineering of plant traits in biotechnology and agriculture.

Covalent Adaptable Poly[2]rotaxane Networks via Dynamic C−N Bond Transalkylation

AbstractCovalent adaptable networks (CANs), a novel class of crosslinked polymers with dynamic covalent bonds, have gained significant attention for combining the durability of thermosets with the reprocessability of thermoplastics, making them promising for emerging applications. Here, we report the first example of poly[2]rotaxane‐type covalent adaptable networks (PRCANs), in which oligo[2]rotaxane backbones characterized by densely packed mechanical bonds, are cross‐linked through dynamic C−N bonds. The oligo[2]rotaxane backbones could guarantee the mechanical properties of the CANs. Under an external force, the synergy of numerous microscopic motions of the cascade [2]rotaxane units, progressively introducing the initially hidden short chains, expands the polymer network, imparting good stretchability to the PRCANs (217 %). On the other hand, the dissociation of host‐guest recognition, followed by the motion of mechanical bonds, constitutes a unique energy dissipation pathway, ultimately enhancing the toughness of PRCANs (7.6 MJ/m3). In contrast, the control CAN, which lacks movable mechanical bonds, demonstrates significantly lower stretchability (40 %) and toughness (1.5 MJ/m3). Moreover, the dynamic C−N bond can undergo high efficiency of 1,2,3‐triazole alkylation and trans‐N‐alkylation exchanges at 1,2,3‐triazolium sites at elevated temperatures, with good reprocessability and without compromising their mechanical performance. This work demonstrates the great potential of oligo[2]rotaxanes as a novel polymer backbone for the development of sustainable materials with excellent mechanical properties.

Gain scheduling of backstepping sliding mode controller using optimized type-1 Sugeno fuzzy logic approach for quadcopter

PurposeThis paper aims to propose an optimized type-1 Sugeno fuzzy logic backstepping sliding mode controller (T1-SFL-BSMC) to control the altitude and rotational behaviour of a quadcopter. The proposed controller improves the performance of the system by overcoming the limitations of conventional trial-and-error techniques for BSMC gain design.Design/methodology/approachIn this study, the type-1 Sugeno fuzzy logic controller (T1-SFLC) technique is proposed which dynamically modifies the BSMC gains. In addition, the parameters of T1-SFLC are further tuned by five different optimization techniques, namely, genetic algorithm, particle swarm optimization (PSO), pattern search (PS), simulated annealing (SA) and adaptive neural network (ANN).FindingsThe effectiveness of these optimization techniques in tuning the T1-SFLCs for the gain scheduled BSMC is evaluated in terms of time response metrics; rising time, settling time and root mean square error (RMSE) for altitude and rotational behaviour. The results demonstrated that PSO-optimized T1-SFL-BSMC provides better responses in terms of time response metrices with reduced rise and settling times and minimized RMSE when compared to other optimized T1-SFL-BSMC.Originality/valueThis paper proposes a novel approach for controlling quadcopters by combining fuzzy logic with a BSMC. It is then improved by using many optimization strategies. A more dependable and effective approach to controller design is provided by the suggested method, which is essential for unmanned aerial vehicle applications.