A centralized framework-based data-driven framework for active distribution system state estimation(DssE)has been widely ***,it is challenged by potential data privacy breaches due to the aggregation of raw measuremen...
详细信息
A centralized framework-based data-driven framework for active distribution system state estimation(DssE)has been widely ***,it is challenged by potential data privacy breaches due to the aggregation of raw measurement data in a data center.A personalized federated learningbased DssE method(PFL-DssE)is proposed in a decentralized training framework for *** validation confirms that PFL-DssE can effectively and efficiently maintain data confidentiality and enhance estimation accuracy.
Currently,the challenge lies in the traditional intelligent algorithm’s ability to effectively address the e-hailing repositioning *** identifying the underlying characteristics in extensive traffic data within a lim...
详细信息
Currently,the challenge lies in the traditional intelligent algorithm’s ability to effectively address the e-hailing repositioning *** identifying the underlying characteristics in extensive traffic data within a limited timeframe is difficult,ultimately preventing the achievement of the most optimal *** paper suggests a hybrid computing architecture involving reinforcement learning and quantum annealing based on intuitive *** reasoning aims to enhance performance in scenarios with poor system robustness,complex tasks,and diverse goals.A deep learning model is constructed,trained to extract scene features,and combined with expert knowledge,then transformed into a quantum annealable *** final strategy is obtained using a D-wave quantum computer with quantum tunneling effect,which helps in finding optimal solutions by jumping out of local suboptimal *** on 400000 real data,four algorithms are compared:minimum-cost flow,sequential markov decision process,hot-dot strategy,and driver-prefer *** average total revenue increases by about 10%and vehicle utilization by about 15%in various *** summary,the proposed architecture effectively solves the e-hailing reposition problem,offering new directions for robust artificial intelligence in big data decision problems.
Due to small size and high occult,metacarpophalangeal fracturediagnosis displays a low accuracy in terms of fracture detection and locationin X-ray *** efficiently detect metacarpophalangeal fractures on Xrayimages as...
详细信息
Due to small size and high occult,metacarpophalangeal fracturediagnosis displays a low accuracy in terms of fracture detection and locationin X-ray *** efficiently detect metacarpophalangeal fractures on Xrayimages as the second opinion for radiologists,we proposed a novel onestageneural network namedMPFracNet based ***,a deformable bottleneck block(DBB)was integrated into the bottleneckto better adapt to the geometric variation of the ***,an integrated feature fusion module(IFFM)was employed to obtain morein-depth semantic and shallow detail ***,Focal Loss andBalanced L1 Loss were introduced to respectively attenuate the imbalancebetween positive and negative classes and the imbalance between detectionand location *** assessed the proposed model on the test set andachieved an AP of 80.4%for the metacarpophalangeal fracture *** estimate the detection performance for fractures with different difficulties,the proposed model was tested on the subsets of metacarpal,phalangeal andtiny fracture test sets and achieved APs of 82.7%,78.5%and 74.9%,*** proposed framework hasstate-of-the-art performance for detectingmetacarpophalangeal fractures,which has a strong potential application valuein practical clinical environments.
Many important problems in science and engineering require solving the so-called parametric partial differential equations(PDEs),i.e.,PDEs with different physical parameters,boundary conditions,shapes of computational...
详细信息
Many important problems in science and engineering require solving the so-called parametric partial differential equations(PDEs),i.e.,PDEs with different physical parameters,boundary conditions,shapes of computational domains,*** reduced order modeling techniques accelerate the solution of the parametric PDEs by projecting them onto a linear trial manifold constructed in the ofline *** methods often need a predefined mesh as well as a series of precomputed solution snapshots,and may struggle to balance between the efficiency and accuracy due to the limitation of the linear *** the nonlinear representation of neural networks(NNs),we propose the Meta-Auto-Decoder(MAD)to construct a nonlinear trial manifold,whose best possible performance is measured theoretically by the decoder *** on the meta-learning concept,the trial manifold can be learned in a mesh-free and unsupervised way during the pre-training *** adaptation to new(possibly heterogeneous)PDE parameters is enabled by searching on this trial manifold,and optionally fine-tuning the trial manifold at the same *** numerical experimentsshow that the MAD method exhibits a faster convergence speed without losing the accuracy than other deep learning-based methods.
High dropout rates globally perpetuate educational disparities with various underlying causes. Despite numerousstrategies to address this issue, more attention should be given to understanding and addressing student ...
详细信息
learning meaningful representations of images in scientific domains that are robust to variations in centroids and orientations remains an important challenge. Here we introduce centroid- and orientation-aware disenta...
详细信息
learning meaningful representations of images in scientific domains that are robust to variations in centroids and orientations remains an important challenge. Here we introduce centroid- and orientation-aware disentangling autoencoder (CODAE), an encoder-decoder-based neural network that learns meaningful content of objects in a latent space. specifically, a combination of a translation- and rotation-equivariant encoder, Euler encoding and an image moment loss enables CODAE to extract features invariant to positions and orientations of objects of interest from randomly translated and rotated images. We evaluate this approach on several publicly available scientific datasets, including protein images from life sciences, four-dimensional scanning transmission electron microscopy data from material science and galaxy images from astronomy. The evaluation shows that CODAE learns centroids, orientations and their invariant features and outputs, as well as aligned reconstructions and the exact view reconstructions of the input images with high quality.
Diagnosing multi-stage diseases typically requires doctors to consider multiple data sources,including clinical symptoms,physical signs,biochemical test results,imaging findings,pathological examination data,and even ...
详细信息
Diagnosing multi-stage diseases typically requires doctors to consider multiple data sources,including clinical symptoms,physical signs,biochemical test results,imaging findings,pathological examination data,and even genetic *** applying machine learning modeling to predict and diagnose multi-stage diseases,several challenges need to be ***,the model needs to handle multimodal data,as the data used by doctors for diagnosis includes image data,natural language data,and structured ***,privacy of patients’data needs to be protected,as these data contain the most sensitive and private ***,considering the practicality of the model,the computational requirementsshould not be too *** address these challenges,this paper proposes a privacy-preserving federated deep learning diagnostic method for multi-stage *** method improves the forward and backward propagation processes of deep neural network modeling algorithms and introduces a homomorphic encryption step to design a federated modeling algorithm without the need for an *** also utilizes dedicated integrated circuits to implement the hardware Paillier algorithm,providing accelerated support for homomorphic encryption in ***,this paper designs and conducts experiments to evaluate the proposed *** experimental resultsshow that in privacy-preserving federated deep learning diagnostic modeling,the method in this paper achieves the same modeling performance as ordinary modeling without privacy protection,and has higher modeling speed compared to similar algorithms.
The process of cultivating soil for crop planting and domesticating animals is known as agriculture. A growing agriculture sector indicates an improving economy. Agriculture is considered as the initial pillar that su...
详细信息
The process of cultivating soil for crop planting and domesticating animals is known as agriculture. A growing agriculture sector indicates an improving economy. Agriculture is considered as the initial pillar that supports global food safety. Additionally, it controls the majority of the global economy. since we depend on agriculture for survival, it needs to be regularly supervised by us. In this global era of computerization, humans depend entirely on cyberspace material as it issuper-fast and takes less time as compared to humans. Hence, human vision can be replicated by computer vision. Visual data and information are processed and analyzed using computer hardware and software. It covers the procedures for gathering, sending, processing, filtering, storing, and comprehending visual data. The study of computational theory can direct computer vision research, and a variety of applications offer a solid foundation and research platform. The use of machine vision has recently increased in response to the growing need for fast and precise ways to track the production of fruit. Machine learning (ML) algorithms make it possible to swiftly and reliably analyze enormous amounts of data, regardless of complexity. It is already widely used in many domains, such as credit analysis, fraud detection, defect sophisticated spam filters, picture recognition patterns, prediction models, and inspection of product features. But with so many options available, it is critical to understand the unique qualities of each approach and the optimal situation in which to apply it. In this review, we have discussed in detail the use of artificial intelligence (AI) in fruit production and summarized more than 110 research applications of AI in fruit production technology. As of now, this review is the first compilation work on the application and prospects of AI-based technology in fruit production systems. This review will provide a single-point comprehensive source of information for acad
Contemporary neural networks frequently encounter the challenge of catastrophic forgetting, wherein newly acquired learning can overwrite and erase previously learned information. The paradigm of continual learning of...
详细信息
The stability problem of power grids has become increasingly serious in recent years as the size of novel power systems *** order to improve and ensure the stable operation of the novel power system,thisstudy propose...
详细信息
The stability problem of power grids has become increasingly serious in recent years as the size of novel power systems *** order to improve and ensure the stable operation of the novel power system,thisstudy proposes an artificial emotional lazy Q-learning method,which combines artificial emotion,lazy learning,and reinforcement learning for static security and stability analysis of power ***,thisstudy compares the analysis results of the proposed method with those of the small disturbance method for a stand-alone power system and verifies that the proposed lazy Q-learning method is able to effectively screen useful data for learning,and improve the static security stability of the new type of power system more effectively than the traditional proportional-integral-differential control and Q-learning methods.
暂无评论