Program path analysis is an essential component of software defect detection and quality assurance. Accurately identifying infeasible paths can prevent false positives caused by invalid paths, enabling developers to pinpoint actual defects more efficiently and enhancing overall software quality and reliability. This paper proposes an integrated approach for determining infeasible paths based on program path features and constraint-based reinforcement learning. First, a loop-structure path search and reduction algorithm is proposed to systematically simplify path explosion induced by loops. Then, a global subgraph-based path reduction algorithm is introduced to effectively remove redundant and irrelevant paths. Subsequently, we propose a path set generation algorithm guided by control and implication relationships to construct an optimized path set. Path constraints and symbolic path constraints are used to enhance semantic representation. Finally, a reinforcement learning-based model utilizing reachability rewards and exploration rewards to dynamically determine path reachability. Experimental results show that our proposed approach significantly reduces path explosion, accurately identifies infeasible paths and outperforms existing methods in terms of accuracy and computational efficiency.

One of the most effective methods for detecting software security vulnerabilities is taint analysis. Some software defects originate from certain external input data. Analyzing the taint sources and the data flow propagation from these sources to defect points through static analysis can help us understand the causes of software defects and reduce the difficulty of debugging them. This paper combines intraprocedural and interprocedural analysis methods to obtain global taint source information. A novel propagation path calculation algorithm is proposed, incorporating predecessor node computation and alias analysis, effectively reducing the negative impact of irrelevant code on the performance of taint analysis. This method not only helps detect errors that lead to vulnerabilities but also analyzes the impact of vulnerable input data on the system. Based on the global taint source analysis algorithm, we developed a static taint source analysis prototype tool for C programs, called AWsTS. Experiments conducted on five open-source projects show that AWsTS improves the accuracy of analysis results without increasing the required analysis time. The average precision for intra-procedural taint source analysis is 93.4%, and the average recall is 90.2%. Similarly, for interprocedural taint source analysis, the average precision is 87.6%, and the average recall is 84.9%. Additionally, AWsTS can output taint propagation paths, providing valuable support for further taint analysis.

Dataset distillation is utilized to condense large datasets into smaller synthetic counterparts, effectively reducing their size while preserving their crucial characteristics. In Federated Learning (FL) scenarios, where individual devices or servers often lack substantial computational power or storage capacity, the use of dataset distillation becomes particularly advantageous for processing large volumes of data efficiently. Current research in dataset distillation for FL has primarily focused on enhancing accuracy and reducing communication complexity, but it has largely neglected the potential risk of backdoor attacks. To solve this issue, in this paper, we propose three adaptive dataset condensation based backdoor attacks against dataset distillation for FL. Adaptive attacks in dataset distillation for FL dynamically modify triggers during the training process. These triggers, embedded in the synthetic data, are designed to bypass traditional security detection. Moreover, these attacks employ self-adaptive perturbations to effectively respond to variations in the model's parameters. Experimental results show that the proposed adaptive attacks achieve at least 5.87% higher success rates, while maintaining almost the same clean test accuracy, compared to three benchmark methods.

Federated Learning is a promising paradigm, of-fering advantages such as access to extensive datasets and robust data privacy preservation. However, a notable challenge arises from the predominant reliance of most federated learning algorithms on labeled data for model training, leaving a limited presence of algorithms capable of effectively utilizing unlabeled data. In real-world scenarios, individuals using electronic devices inadvertently generate copious volumes of unlabeled data, often accompanied by a small fraction of labeled data. This wealth of unlabeled data harbors valuable information, underscoring the importance of developing high-quality federated semi-supervised learning algorithms adept at harnessing both labeled and unlabeled data. This paper introduces FedIDE, an advanced federated semi-supervised learning algorithm. Our approach integrates the principles of pseudo-labeling and instance discrimination, drawing inspiration from contrastive learning, to unlock the potential of unlabeled data. Concurrently, supervised learning is applied to a labeled dataset to enhance model performance. Additionally, we design the LocalEMA local model update algorithm, which amalgamates local and global models during local training, yielding a hybrid model. FedIDE undergoes extensive testing across multiple datasets, surpassing state-of-the-art baselines.

With the advancement of federated crowdsourcing services, the associated privacy concerns have attracted growing attention from both academia and industry. Existing privacy laws impose strict requirements concerning the right to be forgotten for data used in training AI models. In federated crowdsourcing services, the right to be forgotten is guaranteed through federated unlearning. Current federated unlearning solutions encompass a two-step process: first, eliminating model updates associated with the target data to achieve unlearning, followed by retraining among the remaining clients to restore the performance of federated crowdsourcing services. However, this indiscriminate removal of model updates, while safeguarding the privacy of the target data, also greatly undermines the generalization performance of the global model. Moreover, relying on client-side retraining imposes additional economic costs on the federated crowdsourcing service. To tackle the above issues, this paper proposes an efficient federated unlearning framework for federated crowdsourcing services, which is based on adaptive parameter clipping and data-free distillation. We first compute the Fisher information matrix (FIM) to approximate the correlation between the target data and all model parameters, which is utilized to adaptively clip each parameter of the global model. Then, we model the softmax layer of the global model to synthesize pseudo-samples, enabling the retrain process on the crowdsourcing platform for the recovery of generalization performance. We conducted extensive experiments on three datasets, and the results demonstrate that our proposed framework not only possesses outstanding data removal capability but also outperforms the comparison methods in terms of computation time and storage space.

Object detection through deep learning techniques plays a pivotal role in various industrial applications, such as defect detection. With industries increasingly recognizing the importance of protecting sensitive data, there is a growing interest in collaborative detector training using federated learning (FL). However, existing FL solutions face challenges in effectively addressing object detection tasks with limited labeled data across practical institutions. This challenge is especially pronounced in scenarios with densely packed objects, where obtaining sufficient labels is time consuming and costly. In this article, we present an innovative federated semi-supervised learning (SSL) framework expressly designed for object detection in densely packed scenes (FSSLOD) to overcome above challenges. To achieve this, our approach leverages a teacher-student network on the client side for local SSL and employs a designed consistency loss to align the output of the teacher network with that of the student network. Furthermore, we present an elastic update mechanism to mitigate the intricate issue of data distribution disparity by preventing the inclusion of inadequately trained knowledge into the shared model. Comprehensive evaluations on two real-world object detection data sets demonstrate that the proposed method significantly enhances object detection performance in densely packed scenes while also ensuring data privacy.

Lightweight models with high accuracy is critical for edge intelligence. Although the Knowledge Distillation (KD) has been successfully applied to reduce the accuracy loss of quantized neural networks, especially for resource-constrained edge devices, the process of pre-training complex high-precision teacher networks in KD however, will bring huge training overhead. Recently proposed online distillation frameworks offer a good solution for teacher-free distillation, but the regularization effect and simple average aggregation of KD further weaken the representation capability of quantized models that have been reconstructed. In this work, we propose Precision-Mixed and Weight-Average Ensemble (PMWAE) consisting of multiple group members and a group leader. PMWAE provides additional knowledge by changing the bit-precision of the activation and generates aggregated weights for each member in group by attention-based mechanism. The ensemble knowledge is further passed to the group leader to obtain the final model. Extensive experiments on the CIFAR-10/100 and ImageNet-1K datasets show that our method outperforms the existing state-of-the-art methods, both on standard convolutions and depth-wise separable convolutions.

Federated learning (FL) lately has shown much promise in improving the shared model and preserving data privacy. However, these existing methods are only of limited utility in the Internet of Things (IoT) scenarios, as they either heavily depend on high-quality labeled data or only perform well under idealized conditions, which typically cannot be found in practical applications. In this article, we propose a novel federated unsupervised learning method for image classification without the use of any ground truth annotations. In IoT scenarios, a big challenge is that decentralized data among multiple clients is normally nonindependent and identically distributed (non-IID), leading to performance degradation. To address this issue, we further propose a dynamic update mechanism that can decide how to update the local model based on weights divergence. Extensive experiments show that our method outperforms all baseline methods by large margins, including +6.67% on CIFAR-10, +5.15% on STL-10, and +8.44% on SVHN in terms of classification accuracy. In particular, we obtain promising results on Mini-ImageNet and COVID-19 data sets and outperform several federated unsupervised learning methods under non-IID settings.

Federated learning(FL) is a privacy-preserving dis-tributed learning paradigm in which clients cooperate with each other to train a global model. It is becoming progressively prevalent with the rapid development of edge devices. A critical challenge in federated learning is the data heterogeneity among clients, resulting in the global model generated by standard federated learning being unable to be adapted to all clients. To tackle this problem, we propose the CFedPer for personalized FL, which generates a personalized model for each cluster after clustering to address the deficiency of standard federated learning. Our algorithm is organized into two optimization phases. The pre-start phase clusters clients by our proposed similarity-based clustering model using distribution vector and similarity matrix. In the in-training phase, we represent the neural network as the base layer and personalization layer and propose a novel optimization objective with a regularization term for the personalization layer to achieve a balance between per-sonalization and generalization, preventing over-personalization. Extensive experiments on various datasets and data distributions indicate that the performance of our algorithm is superior to the existing algorithms in terms of average local accuracy and variance among clients.

Federated learning (FL) is considered to be a promising paradigm to solve data privacy disclosure in large-scale machine learning. To further enhance the privacy protection of federated learning, prior works incorporate the differentially private data perturbation into the federated system. But it is not feasible given the impairment of the model from noise, as adding Gaussian noise to achieve differential privacy (DP) deteriorates the accuracy of the model. In particular, the assumption that the sophisticated system is homogeneous is not realistic for real scenarios. Heterogeneous networks exacerbate noise disruptions. In this paper, we present FedSeC, a novel differential private federated learning (DP-FL) framework which operates with robust convergence and high-accuracy while achieving adequate privacy protection. FedSeC improves upon naive combinations of federated learning and differential privacy approaches with an updates-based optimization of relative-staleness and semi-synchronous approach for fast convergence in heterogeneous networks. Moreover, we propose a valid client selection scheme to trade-off fair resource allocation and discriminatory incentives. Through extensive experimental validation of our method in three different heterogeneities, we show that FedSeC outperforms the previous state-of-the-art method.

In the Industrial Internet of Things (IIoT), model and computing power sharing among devices can improve resource utilization and work efficiency. However, data privacy and security issues hinder the sharing process. Besides, in the process of model sharing, due to the customization of industrial equipment functions and the high separation of model and task types between devices, it is difficult to share model and optimize models among devices with different task requirements. In this article, we propose an adaptively federated multitask learning (AFL) for IIoT devices efficiently model sharing. Inspired by the parameter sharing mechanism, AFL builds a sparse sharing structure by designing an iterative pruning network and generating subnets for each task. Moreover, for better share relevant information, we further propose tailored task mask layers for effectively training specialized subnets, and an adaptive loss function to dynamically adjust the priority between tasks. Extensive experiments show that AFL can successfully fit hundreds of tasks from different devices into one model, which preserves both high accuracy and system scalability, and outperforms other related approaches that naively combine federated learning with multitask learning.

Federated learning (FL) is a distributed framework for machine learning (ML) model training. Training agents upload local model parameters rather than original training data, and the central server performs parameter aggregation. FL can protect user data privacy and break the information island when training the ML model. Federated Average (FedAvg) is an aggregation method commonly used in the FL training task. The central server calculates the mean value of the local model parameters to obtain the new global parameters. FedAvg assumes that all the training agents are honest, which means the central server lacks terminal agents' knowability. When there are attackers in the training agents, the global model's performance may be deeply affected, and the training task cannot be completed normally. To solve this problem, we propose a Federated Learning method with aggregation based on the Importance of training agent (FedIM), in which the central server performs pre-evaluation on the agent parameters before aggregation, calculates the weights of parameters according to the historical behavior records of training terminals and performs federated aggregation to improve the anti-poisoning ability of learning task. Experiments show that our method can effectively improve the global model's anti-poisoning ability and accelerate the training speed compared with the FedAvg method when malicious agents are involved.

The blockchain system allows various trans-actions and information storage to be executed in a decentralized manner, while smart contracts require multiple nodes to be executed in the local sandbox environment according to preset settings to ensure the consistency of each node, which makes smart contracts unable to proactively obtain data from the outside world. Decentralized oracle can realize the acquisition of off-chain data with a low speed under the premise of ensuring the decentralization of the blockchain. Some oracles use on-chain data storage and maintenance to speed up data acquisition, but this will face higher costs of data storage and maintenance, so current oracles cannot simultaneously ensure privacy and security while taking into account execution cost and processing speed. In this article, we propose Select-Storage, a new oracle design pattern to achieve low operating cost and high processing speed without compromising security. Through experimental analysis, and comparison with other design patterns in processing time and on-chain and off-chain call costs, we have proved the superiority of the Select-Storage design pattern.