Search

Federated learning is a revolutionary deep learning model, and it enables users to train the global model cooperatively without exposing their private data. However, some malicious clients can lead to the risk of single point of failure and privacy disclosure, this poses a serious threat to the security of federated learning. In response to the above issues, based on the existing research, this article proposed a blockchain empowered multi edge federated learning model. Firstly, this paper proposed to use blockchain instead of central server to enhance the stability and reliability of model training process; Secondly, this paper proposed a consensus mechanism based on edge computing to achieve a more efficient consensus process; In addition, incorporating reputation assessment into the federated learning training process can transparently measure the contribution value of each participant and standardize the behavior of work nodes. Finally, comparative experiments show that the scheme can maintain high accuracy in the malicious environment, and can resist higher malicious ratio compared with the traditional federated learning algorithm. [ABSTRACT FROM AUTHOR]

In the context of visual question answering (VQA) tasks, "explainability" refers to the various ways in which researchers can explain why a model works in a given task. The lack of interpretability of some existing VQA models has led to a lack of assurance that the models can be used safely in real-life applications, especially in fields such as autonomous driving and healthcare. This raises ethical and moral issues that hinder their implementation in industry. This paper introduces various implementations for enhancing interpretability in visual question answering tasks and categorizes them into four main categories: image interpretation, text interpretation, multi-modal interpretation, modular interpretation, and graph interpretation. The characteristics of each approach are discussed, and further subdivisions are presented for some of them. Furthermore, the paper presents several VQA datasets that aim to enhance interpretability. These datasets primarily focus on incorporating external knowledge bases and annotating image information to improve interpretability. In summary, this paper provides an overview of existing commonly used interpretable methods for VQA tasks and proposes future research directions based on the identified shortcomings of the current approaches. [ABSTRACT FROM AUTHOR]

Constraint satisfaction problem is one of the most basic NP-complete problems in artificial intelligence. Over the years, with the in-depth study of constraint satisfaction problem, scholars at home and abroad have proposed a variety of case models. RB model is an example of growth domain constraint satisfaction problem with precise phase transition, which is very challenging to solve. In order to find a new efficient algorithm and promote the research of RB model algorithm for constraint satisfiability problem, firstly, this paper analyzed the model development and solving techniques of constraint satisfaction problem. Secondly, it sorted out various examples of RB model solving algorithms, and divided the solving algorithm literature into backtracking heuristic, information propagation and meta-heuristic related improved algorithms. This paper compared and summarized the algorithm principles, improved strategies, convergence and accuracy. Finally, it gave the research trend and development direction of solving RB model example algorithm. [ABSTRACT FROM AUTHOR]

In view of AI model production and training, the traditional script based physical server or cluster mode has problems such as training inference separated, insufficient resource utilization, difficult migration of computing environment, and lengthy training process. This paper proposed a platform architecture for deep learning model training, the architecture divided into four layers: data platform layer, computing platform layer, training suite layer, and management platform layer. Firstly, it proposed an integrated framework masks differences in network structures and optimized the graphs. Secondly, it researched an adaptive resource matching mechanism based on GPU state reduced communication costs. At the same time, it improved re- source utilization by providing a heuristic algorithm based label matching scheduling algorithm. Moreover, the establishment of tenant management and disaster recovery mechanisms ensured the safety and reliability of the system platform. Finally, it established the validation of the usability, safety, reliability, and scalability through the simulation platform. Through the construction of deep learning platform, it will accelerate the implementation of AI production and promote the prosperity and development of AI technology and ecology. [ABSTRACT FROM AUTHOR]

Due to the lack of tailored security and privacy technology in many fields such as Internet of Things, artificial intelligence and big data, the system based on distributed ledger technology (DLT) can well meet their the security requirements as a decentralized data management method. Consequently, systems based on distributed ledger technology emerge in an endless stream. In order to help subsequent researchers understand the field of distributed ledger technology more comprehensively and quickly, this paper reviewed the research results based on distributed ledger technology in recent years. It analyzed the blockchain in depth from the perspectives of data structure, technical principle, consensus mechanism and performance, and compared and analyzed some recent advances in distributed ledger technology in detail based on the current research status, such as Hashgraph, Holochain and Truechain. Finally, this paper summarized the future development prospect of this field. [ABSTRACT FROM AUTHOR]

Aiming at the problem that knowledge tracing research ignores the topological relationship between concepts, thereby limiting the representation of concept states and ultimately affecting the prediction effect, this paper proposed a knowledge state neural reasoning knowledge tracing model. Firstly, it established the concept relationship diagram and concept state diagram. Secondly, it used the diffusion model to obtain the projection of the relationship diagram and the state diagram and complete the fusion. Then, it used the inverse diffusion process to obtain the overall knowledge state representation that integrated the relationship between concepts, and finally predicted the performance of learners. In terms of model effectiveness, comparing four related models on several datasets, the proposed model achieves certain advantages. In terms of interpretability, the proposed model shows the correspondence between the evolution process of knowledge state and the actual answer results. In terms of practical application, the intelligent learning environment based on the model has been applied to an artificial intelligence course and an English grammar course respectively, and has achieved better results than the comparison model. [ABSTRACT FROM AUTHOR]

Message propagation algorithms from statistical physics are widely used in various fields of artificial intelligence, especially in solving combinatorial optimization problems. According to the related literatures of message propagation algorithm, this paper summarized the history of message propagation algorithm and its related application. According to the development of message propagation algorithm, it introduced the concepts of information propagation algorithm for solving the satisfiability problem, which mainly involved the warning propagation algorithm, the belief propagation algorithm and the survey propagation algorithm. This paper described the convergence and effectiveness research of the three kind of algorithms, summarized the application of each algorithm in related fields, and summarized the research path and application direction of message propagation algorithm. [ABSTRACT FROM AUTHOR]

Machine reading comprehension is one of the research hotspots in the field of natural language processing, it is of great significance to improve the reading ability and intelligence level of the machine. In order to follow up the research progress in related fields, this paper reviewed it. Firstly, this paper introduced the development process and main tasks of machine reading comprehension. Secondly, it focused on the current work of multi-choice machine reading comprehension based on deep learning methods, and described the related work from four aspects: semantic matching, pre-training model, semantic reasoning and external knowledge. Next, it summarized the related datasets and evaluat ion metrics . Finally, this paper looked forward to the future development trend of multi-choice machine reading comprehension. [ABSTRACT FROM AUTHOR]

Mild cognitive decline is the early stage of Alzheimer's disease, and the feature extraction and classification of mild cognitive decline using EEG signals is an important method for diagnosing mild cognitive decline. In the automatic detection technology of mild cognitive decline based on electroencephalogram artificial intelligence, the existing research only extracted a certain feature in the electroencephalogram signal or simply concatenates multiple features, which caused these methods to fail to well consider the correlation between different features and it would cause the problem of dimensional disaster. This paper proposed a convolutional neural network based automatic detection algorithm for resting state electroencephalogram data of mild cognitive decline. By extracting the power spectrum and brain network features of the electroencephalogram, it fused the two features by matrix operation, and designed a convolutional neural network to classify the fused features. This method achieves a high accuracy rate on the data set collected by a hospital in Shanghai. In addition, by inputting different subsets of the feature set, this method found the few groups of features that contribute the most to mild cognitive decline, thereby it also had a certain interpretability. Experiments on the dataset of this paper, it p roves the advantages of the power brain network for the automatic diagnosis of mild cognitive decline. [ABSTRACT FROM AUTHOR]

Federated learning and blockchain have a strong commonality, complementarity and compatibility in application fields, architectural features and privacy protection mechanisms. In recent years, some research and applications combined the two technologies and made a lot of progress in terms of data privacy protection, data sharing and computational performance. To help researchers grasp the latest research achievements and development directions, this paper presented a review of blockchain-based federated learning. Firstly, this paper presented the relevant research and the shortcomings of federated learning. Then, it discussed the current research related to blockchain-based federated learning, mainly focused on the architectural features, resources allocation, security mechanism and incentive mechanism. Finally, this paper concluded the future trends and issues of blockchain-based federated learning. [ABSTRACT FROM AUTHOR]

Fuzzy multiple kernel support vector machine (SVM) combines fuzzy SVM with multiple kernel learning (MKL) method which effectively reduces the sensitivity to noises and learning difficulty with the multi-source and heterogeneous data of the traditional SVM model by utilizing membership functions and combinations of multiple kernel functions. Fuzzy multiple kernel SVM has been widely applied in the pattern recognition and artificial intelligence community. This paper summarized the theoretical basis of fuzzy multiple kernel SVM and its current research status. Specifically, this paper were comprehensively reviewed the key problems of the fuzzy multiple kernel SVM, i. e., the design of fuzzy membership functions and MKL methods. Finally, this paper prospected the future research of fuzzy multiple kernel SVM. [ABSTRACT FROM AUTHOR]

The Max-SAT problem is an optimized version of the SAT problem. The goal is to find a set of variable assignments in a given set of clauses so that the maximum number of clauses is satisfied. This problem is a typical NP-hard problem. With the in-depth development of big data and artificial intelligence, the original algorithms in the past are no longer applicable. Designing a new solution algorithm or optimizing the existing solution algorithm is the current research hotspot. Aiming at the limitations of the information dissemination algorithm for solving the random Max-3-SAT problem, this paper proposes a warning dissemination algorithm based on the calculation of variable weights. Combined with the random walk algorithm, a new algorithm WWP+WalkSAT is formed, which is solved by improvement The limitation of, it is better to get a set of effective initial solutions, thereby improving the local search ability of the algorithm. Using some benchmark examples of the Max-SAT International Competition in 2016, the WWP+WalkSAT algorithm and 8 local search algorithms are used for accuracy comparison experiments. The experimental results show that the WWP + walksat algorithm has good performance. [ABSTRACT FROM AUTHOR]

In order to maintain security and reliability of operating system, this paper developed a universal framework for net-work intrusion detection system based on artificial intelligence( AI) model. This framework was capability of detecting all sorts of network traffic, catching out possible malicious connection or attack from Internet and identifying it. Firstly, this system pre-processed the network traffic data by sampling, one-hot encoding, feature selection and normalization for basic feature extrac-tion. Secondly,it utilized feature distribution in network connection instance to build a scoring mechanism for information ex-traction once again. Thirdly, it applied the different types of ML/DL models for different forms of network traffic to do final classification for extracted features. In experiments, this system applied three benchmark datasets, KDDCup99, UNSW-NB15 and CICIDS2017. The result of experiments shows it possesses exceedingly good performance and has better accuracy and F1 score than related works under these datasets. [ABSTRACT FROM AUTHOR]