Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks221 from public data
- Recent Advances for Quantum Neural Networks in Generative Learning143
This paper interprets these QGLMs, covering quantum circuit Born machines, quantum generative adversarial networks, quantum Boltzmann machines, and quantum variational autoencoders, as the quantum extension of classical generative learning models, and explores their intrinsic relations and their fundamental differences.
- GANet: Goal Area Network for Motion Forecasting117
A new goal area-based framework, named Goal Area Network (GANet), for motion forecasting, which models goal areas as preconditions for trajectory prediction, performing more robustly and accurately.
- 68
- XLRS-Bench: Could Your Multimodal LLMs Understand Extremely Large Ultra-High-Resolution Remote Sensing Imagery?65
XLRS-Bench is presented: a comprehensive benchmark for evaluating the perception and reasoning capabilities of MLLMs in ultra-high-resolution RS scenarios, with a primary emphasis on advanced cognitive processes that facilitate real-world decision-making and the capture of spatiotemporal changes.
- Conditional Variational Capsule Network for Open Set Recognition63
This work uses the variational autoencoder framework, with a set of gaussian priors as the approximation for the posterior distribution, to control the compactness of the features of the same class around the center of the gaussians, thus controlling the ability of the classifier in detecting samples from unknown classes.
- User experience model and design strategies for virtual reality-based cultural heritage exhibition52
This study synthesizes three-level coding and quantitative analysis findings from grounded theory, builds a detailed model of the VRCHE UX using theoretical coding, and proposes design strategies.
- 51
- Effect of pulse repetition frequency of high-intensity focused ultrasound on in vitro thrombolysis50
The PRF of delivered HIFU bursts at the same output energy has a significant effect on the thrombi motion, bubble cavitation activities, and subsequently thrombolysis efficiencies.
- A greedy-based stable multi-path routing protocol in mobile ad hoc networks50
A Greedy-based Backup Routing (GBR) protocol is presented that considers both route length and link lifetime to achieve high route stability and has excellent performance in terms of route lifetime, packet delivery ratio, and control overhead.
- 45
- Accuracy Assessment of GlobeLand30 2010 Land Cover over China Based on Geographically and Categorically Stratified Validation Sample Data43
The methods for and results from the first country-wide and statistically based accuracy assessment of GlobeLand30 2010 land cover dataset over China indicate that Globe land cover circa 2010 in China quite accurately, although estimates of accuracy indicators were lower as expected.
- A Lightweight 1-D Convolution Augmented Transformer with Metric Learning for Hyperspectral Image Classification42
This work believes it is the first to combine the metric learning and the transformer model in hyperspectral image classification, and to improve the model classification performance when the available training samples are limited, it uses the 1-D convolution and Mish activation function.
- Debiasing Multimodal Large Language Models via Penalization of Language Priors39
This investigation sheds light on the instability of MLLMs across various decoding configurations, and achieves significant performance improvements-surpassing previously reported results-and raises concerns about the fairness of current evaluation practices.
- Cauchy sparse NMF with manifold regularization: A robust method for hyperspectral unmixing39
This work proposes a robust hyperspectral unmixing model called Cauchy Sparse NMF with manifold regularization (MCSNMF), which incorporates the prior geometric structure into robust subspace learning and outperforms the state-of-the-art NMF based un Mixing methods on both simulated datasets and real scene images.
- 35
- An automatic cervical cell classification model based on improved DenseNet12134
Compared with other state-of-the-art algorithms, the A2SDNet121 model performs better in the multi-classification task of cervical cells, which can significantly improve the accuracy and efficiency of cervical cancer screening.
- 32
- BT Expansion: a Sound and Complete Algorithm for Behavior Planning of Intelligent Robots with Behavior Trees31
This paper proposes BT expansion, an automated planning approach to building intelligent robot behaviors with BTs, and proves the soundness and completeness through the state-space formulation of BTs.
- Mavors: Multi-granularity Video Representation for Multimodal Large Language Model30
Experiments across diverse benchmarks demonstrate Mavors' superiority in maintaining both spatial fidelity and temporal continuity, significantly outperforming existing methods in tasks requiring fine-grained spatio-temporal reasoning.
- Numerical Simulation and Analysis of Fish-Like Robots Swarm30
The results show that at close spacing, the fish in the back can obtain a large wake from the front fish, but suffers large lateral power loss from the lateral fish, and when the spacing is large, both the wake and pressure caused by the front and side fishes become small.
- Learning Capsules for SAR Target Recognition29
A convolutional neural network extension based on Hinton's capsule network is developed to capture spatial relationships specialized in classification between different entities in a SAR image to achieve the accurate and robust classification of SAR images without significantly increasing network complexity.
- MagicFusion: Boosting Text-to-Image Generation Performance by Fusing Diffusion Models28
This work proposes a simple yet effective method called Saliency-aware Noise Blending (SNB), which can empower the fused text-guided diffusion models to achieve more controllable generation by blending the predicted noises of two diffusion models in a saliency-aware manner.
- An image thresholding approach based on Gaussian mixture model27
An image thresholding approach based on Gaussian mixture model (GMM) to solve the problem of noisy and uneven grayscale images by adopting expectation maximization algorithm with a simple initialization strategy and utilizing Bayesian criteria to generate the binary map.
- Tackling Noisy Labels With Network Parameter Additive Decomposition26
To decouple the memorization of clean data and mislabeled data, and further reduce the side effect of mislabeled data, additive decomposition on network parameters is performed.
- 26
- 24
- 23
- Cell-derived biomimetic drug delivery system for inflammatory bowel disease therapy22
A drug delivery system based on bionic technology is expected to achieve a new breakthrough in the targeted therapy of IBD by nanotechnology due to its organic integration of low immunogenicity and natural targeting of biological materials and the controllability and versatility of synthetic nanocarrier design.
- Harnessing Massive Satellite Imagery with Efficient Masked Image Modeling20
A new pre-training pipeline for RS models is presented, featuring the creation of a large-scale RS dataset and an efficient MIM approach, and SelectiveMAE, a pre-training method that dynamically encodes and reconstructs semantically rich patch tokens, thereby reducing the inefficiencies of traditional MIM models caused by redundant background pixels in RS images.
- 20
- D-FusionNet: road extraction from remote sensing images using dilated convolutional block19
D-FusionNet integrates the Dilated Convolutional Block (DCB) module, which serves as a technique for expanding the receptive field and mitigating feature loss, resembling a residual mechanism during encoding, and performs well in road extraction tasks.
- ALLIANCE‐ROS: A Software Framework on ROS for Fault‐Tolerant and Cooperative Mobile Robots19
This work proposes a software framework named ALLIANCE-ROS for developing fault-tolerant cooperative multi-robot systems with abundant software resources available, and demonstrates three application cases including an autonomous roving robot, a security patrol robot and multiple patrol robots.
- Sequential Fusion Based Multi-Granularity Consistency for Space-Time Transformer Tracking17
This work proposes a Space-Time Consistent Transformer Tracker (STCFormer), which uses a sequential fusion framework with multi-granularity consistency constraints to learn spatiotemporal context information and designs three space-time consistent constraints.
- Null-text Guidance in Diffusion Models is Secretly a Cartoon-style Creator17
It is demonstrated that null-text guidance in diffusion models is secretly a cartoon-style creator, i.e., the generated images can be efficiently transformed into cartoons by simply perturbing the null- Text guidance in the sampling process.
- Deep parameter-free attention hashing for image retrieval17
A novel deep parameter-free attention hashing (DPFAH) is proposed to solve problems of insufficiency feature extraction when existing deep hashing method uses the convolutional neural network to extract images semantic features and experiments show that DPFAH method achieves better performance.
- 17
- 16
- The lower He-sea points playing a significant role in postoperative ileus in colorectal cancer treated with acupuncture: based on machine-learning16
Meta-analysis indicates acupuncture’s safety and superior effectiveness over postoperative care alone in facilitating gastrointestinal recovery and machine-learning approaches highlight the importance of the lower He-sea points, including Zusanli (ST36) and Shangjuxu (ST37), in treating POI in CRC patients.
- 16
- Heterogeneous Pseudo-Supervised Learning for Few-shot Person Re-Identification15
A simple yet effective Heterogeneous Pseudo-Supervised Learning (H-PSL) framework based on classical PSL to implement asynchronous match, which boosts the feature expression and then a better label prediction in the following and a knowledge fusion strategy to integrate the pseudo labels and their confidence which are easily obtained by the base model and isomer.
- Integrating Decision Sharing with Prediction in Decentralized Planning for Multi-Agent Coordination under Uncertainty15
This paper presents a novel planning algorithm by combining decision sharing and prediction based on decentralized Monte Carlo Tree Search called Dec-MCTS-SP, which can reduce communication cost significantly while still achieving a surprisingly higher coordination performance.
- CMTG: A Content-Based Mobile Tendency Geocast Routing Protocol in Urban Vehicular Networks15
This paper proposes a content-based mobile tendency geocast (CMTG) routing protocol for urban vehicular networks from the delay tolerant networks (DTNs) perspective to better deal with the high mobility and transient connectivity issues.
- 14
- 13
- 12
- Pairwise Similarity Regularization for Adversarial Domain Adaptation12
A Pairwise Similarity Regularization (PSR) approach that exploits cluster structures of the target domain data and minimizes the divergence between the pairwise similarity of clustering partition and that of pseudo predictions to eliminate the negative effect of unreliable pseudo labels.
- 12
- The convergence and termination criterion of quantum-inspired evolutionary neural networks12
An appropriate termination criterion based on the average convergence rate (ACR) can duly stop the training process of QENN and overcome the limitations of the termination criterionbased on the probability of generating the best solution (PBS).
- 12
- Improving Route Stability in Mobile Ad hoc Networks Based on Link Lifetime12
Both theoretical analyses and simulation results show that LBR has an improvement in route lifetime, packet delivery ratio, end-to-end delay, and so on, compared with single-path routing protocols.
- Deep Hash with Improved Dual Attention for Image Retrieval11
The improved dual attention mechanism (IDA) based on the ResNet18 pre-trained module to extract the feature information of the image, which consists of the position attention module and the channel attention module is introduced.
- An Environmental Perception Framework for Robotic Fish Formation Based on Machine Learning Methods11
A predictive model of flow velocity and a judgement model of spacing between individual platforms for robotic fish formation through monitoring sensors on robotic fish surface are built by methods of polynomial fitting and neural networks based on Computational Fluid Dynamics (CFD) simulation.
- An Explorative Study on Estimating Local Accuracies in Land-Cover Information Using Logistic Regression and Class-Heterogeneity-Stratified Data11
The proposed double-stratification method’s performance and sensitivity to sample size regarding local accuracy estimation in comparison with those of existing methods are investigated through a case study concerning Globeland30 2010 land cover over Wuhan, China.
- Role of Nonmonotonic Constitutive Curves in Extrusion Instabilities11
It could be concluded that the nonmonotonic constitutive curve plays an essential role in extrusion instabilities and the results imply that the die exit singularity may generate or magnify oscillations.
- 11
- Your Neighbor Matters: Towards Fair Decisions Under Networked Interference10
This work introduces "Interference-aware Fairness" (IAF) by defining two forms of discrimination as Self-Fairness (SF) and Peer-Fairness (PF), leveraging advances in interference analysis within causal inference.
- 10
- 10
- Boosting Few-shot Object Detection with Discriminative Representation and Class Margin10
This work proposes a weight transfer strategy to promote model transferability and an attention-based feature enhancement mechanism to learn more robust object proposal feature representations and introduces an angle-guided additive margin classifier to augment instance-level inter-class difference and intra-class compactness.
- 10
- Geostatistical characterization of local accuracies in remotely sensed land cover change categorization with complexly configured reference samples10
A geostatistical method for predictive mapping of local accuracies in any land-cover change maps created properly through adaptive use of complexly configured training sample data is proposed and generally more accurate than Product and should be pursued when enough collocated data are available for estimating cross-correlation.
- 10
- HBTP: Heuristic Behavior Tree Planning with Large Language Model Reasoning9
This paper introduces Heuristic Behavior Tree Planning (HBTP), a reliable and efficient framework for BT generation and proposes methods to address the inaccuracies of LLM reasoning, including action space pruning and reflective feedback, to further enhance both reasoning accuracy and planning efficiency.
- Impact of digital infrastructure on urban green eco-efficiency: an empirical analysis in China9
A comprehensive analysis of 267 prefecture-level cities that have implemented the “Broadband China” strategy as empirical research objects provides valuable insights on balancing green ecological efficiency with urban digitization, offering practical guidance for sustainable development.
- 9
- 9
- 9
- 9
- 9
- A Theoretical Understanding of Neural Network Compression from Sparse Linear Approximation9
This work proposes to use the sparsity-sensitive $\ell_q$-norm ($0<q<1$) to characterize compressibility and provide a relationship between soft sparsity of the weights in the network and the degree of compression with a controlled accuracy degradation bound, and develops adaptive algorithms for pruning each neuron in thenetwork informed by the theory.
- Enhancing Soluble Expression of Phospholipase B for Efficient Catalytic Synthesis of L-Alpha-Glycerylphosphorylcholine9
The proposed cost-effective PLB production and an environmentally friendly PLB-catalyzed system offer a candidate strategy for the industrial production of L-α-GPC.
- 9
- 9
- 8
- 8
- 8
- Does Confusion Really Hurt Novel Class Discovery?7
It is found that NUSA will empirically degrade existing NCD methods if taking no care of sampling errors, and an effective solution is proposed, named hidden-prototype-based discovery network (HPDN), which tries to obtain relatively clean data representations even with the confusedly sampled data.
- 7
- Scaling Few-Shot Learning for the Open World7
This work proposes a new problem named Few-Shot Learning with Many Novel Classes (FSL-MNC), substantially enlarging the number of novel classes, exceeding the count in the traditional FSL setup by over 500-fold and proposes a Simple Hierarchy Pipeline (SHA-Pipeline), a lightweight training strategy to reduce the overhead brought by much more novel classes.
- Robust Normalized Squares Maximization for Unsupervised Domain Adaptation7
A novel normalized squares maximization (NSM) loss in which the maximum squares is normalized by the sum of squares of class sizes to explicitly increase the diversity of UDA predictions.
- 7
- 6
- Repairing Deep Neural Networks Based on Behavior Imitation6
A behavior-imitation based repair framework, BIRDNN, which integrates the two repair paradigms for the first time and can successfully repair buggy DNNs with significantly higher efficiency than state-of-the-art repair tools.
- 6
- 6
- Person Re-Identification Based on Multi-Parts of Local Feature Network6
This paper proposes a novel, simple and efficient person re-id network called MPLFN, which combines two tasks: the classification task and the metric learning task, and achieves a better performance on three mainstream people re-identification datasets.
- 6
- 6
- Transformer-Based Spatial-Temporal Counterfactual Outcomes Estimation5
A novel framework for estimating counterfactual outcomes with spatial-temporal attributes using the Transformer, exhibiting stronger estimation ability and under mild assumptions, the proposed estimator within this framework is consistent and asymptotically normal.
- Exploring the relationship between electrophysiological measures of the electrically evoked auditory brainstem response and speech perception outcomes post-cochlear implantation5
An apex-to-base gradient in EABR responsiveness following prolonged CI use is identified and the threshold and I/O slopes of EABR appear to be informative predictors of speech perception performance in CI users, especially in the low-to-middle frequency range.
- Role‐based attention in deep reinforcement learning for games5
A role‐based attention model for reinforcement learning that uses convolutional neural networks to generate soft attention maps, adding crucial role information in the task, forcing the agent to focus on important features and distinguish task‐related information.
- Deep supervised multimodal semantic autoencoder for cross‐modal retrieval5
This paper presents a three‐stage cross‐modal retrieval method, named MMCA‐CMR, which significantly outperforms state‐of‐the‐art on four widely‐used benchmark datasets and promotes a more reliable cross‐ modal retrieval by learning feature vectors from different modalities and content information simultaneously.
- MulAttenRec: A Multi-level Attention-Based Model for Recommendation5
A multi-level attention mechanism to explore the usefulness of reviews and the significance of words and propose a Multi-level Attention-based Model (MulAttRec) for the recommendation.
- Multi-feature Fusion for Deep Reinforcement Learning: Sequential Control of Mobile Robots5
With more accurate and stable states extracted from the autoencoder, the proposed multi-feature fusion framework improves not only the convergence rate of training phase but also the testing performance of the mobile robot.
- 5
- 5
- Learning Feasible Causal Algorithmic Recourse: A Prior Structural Knowledge Free Approach4
This work aims to develop a novel framework for causal algorithmic recourse that does not rely on neither prior causal graph or SCM knowledge, and proposes to approximate and constrain the variation of the perturbed components, i.e., the exogenous noise variables, by formulating the generation of AR as the structure-preserving intervention.
- 4
- BIRDNN: Behavior-Imitation Based Repair for Deep Neural Networks4
BIRDNN is introduced, a behavior-imitation based DNN repair framework that supports alternative retraining and fine-tuning repair paradigms for DRPs and features more excellent effectiveness, efficiency, and compatibility in repairing buggy DNNs comprehensively compared with state-of-the-art repair methods.
- 4
- Dec-SGTS: Decentralized Sub-Goal Tree Search for Multi-Agent Coordination4
Dec-SGTS (Decentralized Sub-Goal Tree Search) is presented to solve decentralized online multi-agent planning hierarchically and efficiently and achieves much better reward while enjoying a significant reduction of planning time and communication cost compared with Dec-MCTS.
- One‐shot video‐based person re‐identification with variance subsampling algorithm4
A novel sampling strategy to improve the quality of assigned pseudo‐labels, thus promoting the final performance and evaluating the approach on two publicly available datasets, MARS and DukeMTMC‐VideoReID, and achieve state‐of‐the‐art one‐shot performance.
- 4
- 4
- 4
- 3
- Attribute-formed Class-specific Concept Space: Endowing Language Bottleneck Model with Better Interpretability and Scalability3
The Attribute-formed Language Bottleneck Model (ALBM) is proposed, which organizes concepts in the attribute-formed class-specific space, where concepts are descriptions of specific attributes for specific classes.
- 3
- Acupuncture’s Emergence as A Promising Non-Pharmacological Therapy for Appetite Management in Cancer Chemotherapy3
Acupuncture holds promise as a potential non-pharmacological approach for managing anorexia during cancer chemotherapy and well-designed Randomized Controlled Trials (RCTs) with larger participant cohorts, and consistent core outcome measures are essential.
- Auction-Based Behavior Tree Evolution for Heterogeneous Multi-Agent Systems3
A novel two-layer framework designed to learn BTs for heterogeneous MASs that demonstrates robust effectiveness in covering high-priority tasks, thereby validating the efficacy of employing an auction algorithm for generating BTs tailored for heterogeneous MAS.
- 3
- Lightweight target speaker separation network based on joint training3
A lightweight target speaker separation network based on long-short-term memory (LSTM) is proposed, which can reduce the model size and computational delay while maintaining the separation performance and joint loss functions based on speaker registration and speaker separation are proposed for joint training of the network to enhance the model’s performance.
- 3
- 3
- 3
- Improved Localization Algorithm Based on Received Signal Strength Indicator3
A novel 1- dimensional empirical model location algorithm —1-DEMRSSI is proposed for 1-dimensional wireless sensor networks applications such as road surveillance and pipeline transmission and it has been observed that the algorithm is well flexible and practical for linear applications.
- 2
- 2
- 2
- 2
- 2
- 2
- DBTN: An adaptive neural network for multiple-disease detection via imbalanced medical images distribution2
A novel paradigm for multiple disease detection via medical images that combined a meta-heuristic algorithm and transfer learning and adaptively adjust the feature extraction ability by reshaping a few network layers to accommodate the data distribution differences between source and target datasets is proposed.
- Qualitative and Quantitative Model Checking Against Recurrent Neural Networks2
This work presents an extension of linear-time temporal logic to reason about properties with respect to RNNs, such as local robustness, reachability, and some temporal properties, and formalizes the verification obligation as a Hoare-like triple from both qualitative and quantitative perspectives.
- 2
- 2
- 2
- 2
- 2
- Corrigendum: Landscape and predictions of inflammatory bowel disease in China: China will enter the Compounding Prevalence stage around 20302
This research presents a novel probabilistic procedure that allows for direct measurement of the response of the immune system to earthquake-triggered landsliding.
- 2
- 2
- 2
- UMine: Study on Prevalent Co-locations Mining from Uncertain Data Sets2
A novel system called UMine is proposed to compare this two different definitions of the prevalent co-location pattern under uncertain environments with a user-friendly interface and the experimental results show that the difference between these two definitions’ result sets changes as the threshold changes.
- 2
- 2
- 2
- Buffer Management Algorithm of Wireless Sensor Network Based on Time Correlation2
A new buffer management algorithm of wireless sensor network is proposed combining the concepts of active queue management and time correlation that is effective to discard time-related data and reduce redundant information.
- 2
- 2
- A visual state space Model-Based Cross-Domain adaptive detection method for imbalanced medical image distribution1
A new cross-domain adaptive detection method for imbalanced medical images, named medical adaptive mamba detection (Med-AMamDa), which comprises an adaptive selector that uses a sigmoid function to emphasize task-relevant features based on input characteristics and a dual-branch block, termed the combined convolution and state space model (CoCoS) block, which captures local details and global context to enhance detection accuracy.
- 1
- 1
- 1
- 1
- 1
- 1
- Dual-energy computed tomography is a superior diagnostic performance for locating parathyroid tumors1
- Extension of nonnegative constraints to difference-map algorithm on both real and imaginary parts of real space in phase retrieval1
This study proposes the use of nonnegative constraints, reducing the number of unknown variables by half, on both the real and imaginary parts of the object space to reconstruct the image of a complex-valued object in the iterative process of the difference-map algorithm.
- PEPC: Parallel and Extensible PC Implementation for Causal Structure Learning1
This paper proposes an efficient CPU-based parallel implementation of the PC algorithm, called PEPC, which has an extensible framework, and some optimizations are adopted to reduce the CSL computation of thePC algorithm.
- 1
- Characterizing Uncertainty and Enhancing Utility in Remotely Sensed Land Cover Using Error Matrices Localized in Canonical Correspondence Analysis Ordination Space1
The proposed CCAErrMat method is recommendable for characterizing uncertainty and enhancing utilities in land cover maps by analyzing locally constrained error matrices and is also cost-effective in terms of reference sample data.
- Credit assignment for trained neural networks based on Koopman operator theory1
An alternative perspective of linear dynamics on dealing with the credit assignment problem for trained neural networks is presented, based on Koopman operator theory, and a metric with algebraic interpretability for the credit assignments of each network component is defined.
- 1
- 1
- 1
- 1
- 1
- 1
- Multi-Agent Online Coordination via Bayesian Inverse Planning1
This paper proposes an online multi- agent coordination mechanism for each agent in the absence of communication and explores the benefits of Bayesian inverse planning in achieving the Nash equilibrium which provides an optimal coordination solution for each participant in multi-agent environment.
- Refining Land-Cover Maps Based on Probabilistic Re-Classification in CCA Ordination Space1
In conclusion, on condition of refinability, map refinement is well worth pursuing, as it increases accuracy of existing maps, extends utility of reference data, facilitates uncertainty-informed map representation, and enhances the authors' understanding about relationships between reference data and map data.
- 1
- 1
- 1
- Modified Neutral Models as Benchmarks to Evaluate the Dynamics of Land System (DLS) Model Performance1
Using the modified neutral model results as benchmarks, a new proposed land use model, the Dynamics of Land System (DLS) model, for Jilin Province of China from 2003 to 2013 is assessed using the Kappa statistic and Kappain-out statistic for simulation accuracy.
- 1
- 1
- 1
- A Novel Network Traffic Analysis Method Based on Fuzzy Association Rules1
The method sets up the fuzzy association rules and could analyze the traffic of the global network by using data mining algorithm and can represent the traffic’s characters much more precisely and forecast the behaviors of traffic in advance.
- –
- –
- –
- CODE: Cross-Context Identity Extraction for Value Decomposition Based Multi-Agent Reinforcement Learning–
CODE is proposed, a generic identity-aware framework designed to enhance the cross-context generalization capability of value decomposition (VD) based MARL algorithms such as VDN, QMIX and QPLEX, strengthening the contextual adaptability of VD MARL algorithms.
- Enhanced Semantic Alignment in Transformer Tracking via Position Learning and Force-Directed Attention–
An Enhanced Semantic Alignment Transformer Tracker (ESAT) is proposed, which incorporates a repulsion-attraction mechanism applied to the self-attention module, named Force-Directed Attention Semantic Alignment Module (FDA-SAM), adjusting attention scores and enhancing semantic relationships among features.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Towards robust neural networks via a global and monotonically decreasing robustness training strategy–
This paper proposes a global and monotonically decreasing robustness training strategy that takes multiple perturbations into account during each training epoch, and the corresponding robustness losses are combined with monotonally decreasing weights.
- –
- –
- –
- Study on molecular mechanism of Jiufen powder in the treatment of rheumatoid arthritis based on network pharmacology and molecular docking–
Through multi-target and multi-pathway analysis, the potential mechanism of Jiufen Powder in treating rheumatoid arthritis was preliminarily explored, which provided scientific basis for the clinical application and in-depth study of the Chinese medicine.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Construction of Monitoring System and Data platform for Students’ Physical Fitness–
The high accuracy of the measuring system for physical activity and heart rate can measure and evaluate exercise heart rate and physical activity, and can guarantee exercise safety for students and provide supervision and instruction for school physical education and students’ physical health.
- –
- –
- –
- Study of network architecture for Unmanned Aerial Vehicle–
The structure of self-organizing network is analyzed and then an across-layer network architecture for UAV formation and the requirements of formation and external communication are proposed.
- –
- –
- –
- –
- –
- –
- Transdermal drug delivery using microneedle array of sinomenine hydrogel patch.–
When transdermal drug delivery by using microneedle array-hydrogel patch is investigated, the permeation rates increase significantly and the needle-shape,skin pretreating time, and force of themicroneedles play the important roles in the transder mal drug delivery.
- AODV protocol of improving local repair mechanism in MANET–
Simulation results show that the improved AODV protocol is effective in reducing the route spending and point to point average data transfer delay, and it also proves the protocol efficiency and the performance of whole network.
- A Novel Distributed DDoS Attack Defensive System–
The paper studies distributed denial of service attack and proposes a novel defensivs system against DDoS by using parallel and distributed systems that is effective to defense DDoS attack by correspondence of this system′s module.
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-14. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.