arXiv subjects

Hai Dong

Publications and source records attributed to Hai Dong.

At least 37 records · Page 2Linked to original sources

arXiv2024

FLrce: Resource-Efficient Federated Learning with Early-Stopping Strategy

Federated Learning (FL) achieves great popularity in the Internet of Things (IoT) as a powerful interface to offer intelligent services to customers while maintaining data privacy. Under the orchestration of a server, edge devices (also called clients in FL) collaboratively train a global deep-learning model without sharing any local data. Nevertheless, the unequal training contributions among clients have made FL vulnerable, as clients with heavily biased datasets can easily compromise FL by sending malicious or heavily biased parameter updates. Furthermore, the resource shortage issue of the network also becomes a bottleneck. Due to overwhelming computation overheads generated by training deep-learning models on edge devices, and significant communication overheads for transmitting deep-learning models across the network, enormous amounts of resources are consumed in the FL process. This encompasses computation resources like energy and communication resources like bandwidth. To comprehensively address these challenges, in this paper, we present FLrce, an efficient FL framework with a relationship-based client selection and early-stopping strategy. FLrce accelerates the FL process by selecting clients with more significant effects, enabling the global model to converge to a high accuracy in fewer rounds. FLrce also leverages an early stopping mechanism that terminates FL in advance to save communication and computation resources. Experiment results show that, compared with existing efficient FL frameworks, FLrce improves the computation and communication efficiency by at least 30% and 43% respectively.

cs.LG↗

arXiv2024

FEDQ-Trust: Efficient Data-Driven Trust Prediction for Mobile Edge-Based IoT Systems

We introduce FEDQ-Trust, an innovative data-driven trust prediction approach designed for mobile edge-based Internet of Things (IoT) environments. The decentralized nature of mobile edge environments introduces challenges due to variations in data distribution, impacting the accuracy and training efficiency of existing distributed data-driven trust prediction models. FEDQ-Trust effectively tackles the statistical heterogeneity challenges by integrating Federated Expectation-Maximization with Deep Q Networks. Federated Expectation-Maximization's robust handling of statistical heterogeneity significantly enhances trust prediction accuracy. Meanwhile, Deep Q Networks streamlines the model training process, efficiently reducing the number of training clients while maintaining model performance. We conducted a suite of experiments within simulated MEC-based IoT settings by leveraging two real-world IoT datasets. The experimental results demonstrate that our model achieved a significant convergence time reduction of 97% to 99% while ensuring a notable improvement of 8% to 14% in accuracy compared to state-of-the-art models.

cs.DC↗

arXiv2024

RITFIS: Robust input testing framework for LLMs-based intelligent software

The dependence of Natural Language Processing (NLP) intelligent software on Large Language Models (LLMs) is increasingly prominent, underscoring the necessity for robustness testing. Current testing methods focus solely on the robustness of LLM-based software to prompts. Given the complexity and diversity of real-world inputs, studying the robustness of LLMbased software in handling comprehensive inputs (including prompts and examples) is crucial for a thorough understanding of its performance. To this end, this paper introduces RITFIS, a Robust Input Testing Framework for LLM-based Intelligent Software. To our knowledge, RITFIS is the first framework designed to assess the robustness of LLM-based intelligent software against natural language inputs. This framework, based on given threat models and prompts, primarily defines the testing process as a combinatorial optimization problem. Successful test cases are determined by a goal function, creating a transformation space for the original examples through perturbation means, and employing a series of search methods to filter cases that meet both the testing objectives and language constraints. RITFIS, with its modular design, offers a comprehensive method for evaluating the robustness of LLMbased intelligent software. RITFIS adapts 17 automated testing methods, originally designed for Deep Neural Network (DNN)-based intelligent software, to the LLM-based software testing scenario. It demonstrates the effectiveness of RITFIS in evaluating LLM-based intelligent software through empirical validation. However, existing methods generally have limitations, especially when dealing with lengthy texts and structurally complex threat models. Therefore, we conducted a comprehensive analysis based on five metrics and provided insightful testing method optimization strategies, benefiting both researchers and everyday users.

cs.SE↗

arXiv2023

LEAP: Efficient and Automated Test Method for NLP Software

The widespread adoption of DNNs in NLP software has highlighted the need for robustness. Researchers proposed various automatic testing techniques for adversarial test cases. However, existing methods suffer from two limitations: weak error-discovering capabilities, with success rates ranging from 0% to 24.6% for BERT-based NLP software, and time inefficiency, taking 177.8s to 205.28s per test case, making them challenging for time-constrained scenarios. To address these issues, this paper proposes LEAP, an automated test method that uses LEvy flight-based Adaptive Particle swarm optimization integrated with textual features to generate adversarial test cases. Specifically, we adopt Levy flight for population initialization to increase the diversity of generated test cases. We also design an inertial weight adaptive update operator to improve the efficiency of LEAP's global optimization of high-dimensional text examples and a mutation operator based on the greedy strategy to reduce the search time. We conducted a series of experiments to validate LEAP's ability to test NLP software and found that the average success rate of LEAP in generating adversarial test cases is 79.1%, which is 6.1% higher than the next best approach (PSOattack). While ensuring high success rates, LEAP significantly reduces time overhead by up to 147.6s compared to other heuristic-based methods. Additionally, the experimental results demonstrate that LEAP can generate more transferable test cases and significantly enhance the robustness of DNN-based systems.

cs.SE↗

arXiv2023

Towards Stealthy Backdoor Attacks against Speech Recognition via Elements of Sound

Deep neural networks (DNNs) have been widely and successfully adopted and deployed in various applications of speech recognition. Recently, a few works revealed that these models are vulnerable to backdoor attacks, where the adversaries can implant malicious prediction behaviors into victim models by poisoning their training process. In this paper, we revisit poison-only backdoor attacks against speech recognition. We reveal that existing methods are not stealthy since their trigger patterns are perceptible to humans or machine detection. This limitation is mostly because their trigger patterns are simple noises or separable and distinctive clips. Motivated by these findings, we propose to exploit elements of sound ($e.g.$, pitch and timbre) to design more stealthy yet effective poison-only backdoor attacks. Specifically, we insert a short-duration high-pitched signal as the trigger and increase the pitch of remaining audio clips to `mask' it for designing stealthy pitch-based triggers. We manipulate timbre features of victim audios to design the stealthy timbre-based attack and design a voiceprint selection module to facilitate the multi-backdoor attack. Our attacks can generate more `natural' poisoned samples and therefore are more stealthy. Extensive experiments are conducted on benchmark datasets, which verify the effectiveness of our attacks under different settings ($e.g.$, all-to-one, all-to-all, clean-label, physical, and multi-backdoor settings) and their stealthiness. The code for reproducing main experiments are available at \url{https://github.com/HanboCai/BadSpeech_SoE}.

cs.SD↗

arXiv2023

RPDP: An Efficient Data Placement based on Residual Performance for P2P Storage Systems

Storage systems using Peer-to-Peer (P2P) architecture are an alternative to the traditional client-server systems. They offer better scalability and fault tolerance while at the same time eliminate the single point of failure. The nature of P2P storage systems (which consist of heterogeneous nodes) introduce however data placement challenges that create implementation trade-offs (e.g., between performance and scalability). Existing Kademlia-based DHT data placement method stores data at closest node, where the distance is measured by bit-wise XOR operation between data and a given node. This approach is highly scalable because it does not require global knowledge for placing data nor for the data retrieval. It does not however consider the heterogeneous performance of the nodes, which can result in imbalanced resource usage affecting the overall latency of the system. Other works implement criteria-based selection that addresses heterogeneity of nodes, however often cause subsequent data retrieval to require global knowledge of where the data stored. This paper introduces Residual Performance-based Data Placement (RPDP), a novel data placement method based on dynamic temporal residual performance of data nodes. RPDP places data to most appropriate selected nodes based on their throughput and latency with the aim to achieve lower overall latency by balancing data distribution with respect to the individual performance of nodes. RPDP relies on Kademlia-based DHT with modified data structure to allow data subsequently retrieved without the need of global knowledge. The experimental results indicate that RPDP reduces the overall latency of the baseline Kademlia-based P2P storage system (by 4.87%) and it also reduces the variance of latency among the nodes, with minimal impact to the data retrieval complexity.

cs.DC↗

arXiv2023

AICons: An AI-Enabled Consensus Algorithm Driven by Energy Preservation and Fairness

Blockchain has been used in several domains. However, this technology still has major limitations that are largely related to one of its core components, namely the consensus protocol/algorithm. Several solutions have been proposed in literature and some of them are based on the use of Machine Learning (ML) methods. The ML-based consensus algorithms usually waste the work done by the (contributing/participating) nodes, as only winners' ML models are considered/used, resulting in low energy efficiency. To reduce energy waste and improve scalability, this paper proposes an AI-enabled consensus algorithm (named AICons) driven by energy preservation and fairness of rewarding nodes based on their contribution. In particular, the local ML models trained by all nodes are utilised to generate a global ML model for selecting winners, which reduces energy waste. Considering the fairness of the rewards, we innovatively designed a utility function for the Shapley value evaluation equation to evaluate the contribution of each node from three aspects, namely ML model accuracy, energy consumption, and network bandwidth. The three aspects are combined into a single Shapley value to reflect the contribution of each node in a blockchain system. Extensive experiments were carried out to evaluate fairness, scalability, and profitability of the proposed solution. In particular, AICons has an evenly distributed reward-contribution ratio across nodes, handling 38.4 more transactions per second, and allowing nodes to get more profit to support a bigger network than the state-of-the-art schemes.

cs.DC↗

arXiv2022

VSVC: Backdoor attack against Keyword Spotting based on Voiceprint Selection and Voice Conversion

Keyword spotting (KWS) based on deep neural networks (DNNs) has achieved massive success in voice control scenarios. However, training of such DNN-based KWS systems often requires significant data and hardware resources. Manufacturers often entrust this process to a third-party platform. This makes the training process uncontrollable, where attackers can implant backdoors in the model by manipulating third-party training data. An effective backdoor attack can force the model to make specified judgments under certain conditions, i.e., triggers. In this paper, we design a backdoor attack scheme based on Voiceprint Selection and Voice Conversion, abbreviated as VSVC. Experimental results demonstrated that VSVC is feasible to achieve an average attack success rate close to 97% in four victim models when poisoning less than 1% of the training data.

cs.SD↗

arXiv2022

PBSM: Backdoor attack against Keyword spotting based on pitch boosting and sound masking

Keyword spotting (KWS) has been widely used in various speech control scenarios. The training of KWS is usually based on deep neural networks and requires a large amount of data. Manufacturers often use third-party data to train KWS. However, deep neural networks are not sufficiently interpretable to manufacturers, and attackers can manipulate third-party training data to plant backdoors during the model training. An effective backdoor attack can force the model to make specified judgments under certain conditions, i.e., triggers. In this paper, we design a backdoor attack scheme based on Pitch Boosting and Sound Masking for KWS, called PBSM. Experimental results demonstrated that PBSM is feasible to achieve an average attack success rate close to 90% in three victim models when poisoning less than 1% of the training data.

cs.SD↗

arXiv2022

Resolution of 100 photons and quantum generation of unbiased random numbers

Macroscopic quantum phenomena, such as observed in superfluids and superconductors, have led to promising technological advancements and some of the most important tests of fundamental physics. At present, quantum detection of light is mostly relegated to the microscale, where avalanche photodiodes are very sensitive to distinguishing single-photon events from vacuum but cannot differentiate between larger photon-number events. Beyond this, the ability to perform measurements to resolve photon numbers is highly desirable for a variety of quantum information applications including computation, sensing, and cryptography. True photon-number resolving detectors do exist, but they are currently limited to the ability to resolve on the order of 10 photons, which is too small for several quantum state generation methods based on heralded detection. In this work, we extend photon measurement into the mesoscopic regime by implementing a detection scheme based on multiplexing highly quantum-efficient transition-edge sensors to accurately resolve photon numbers between zero and 100. We then demonstrate the use of our system by implementing a quantum random number generator with no inherent bias. This method is based on sampling a coherent state in the photon-number basis and is robust against environmental noise, phase and amplitude fluctuations in the laser, loss and detector inefficiency as well as eavesdropping. Beyond true random number generation, our detection scheme serves as a means to implement quantum measurement and engineering techniques valuable for photonic quantum information processing.

quant-ph↗

arXiv2022

Deep Edge Intelligence: Architecture, Key Features, Enabling Technologies and Challenges

With the breakthroughs in Deep Learning, recent years have witnessed a massive surge in Artificial Intelligence applications and services. Meanwhile, the rapid advances in Mobile Computing and Internet of Things has also given rise to billions of mobile and smart sensing devices connected to the Internet, generating zettabytes of data at the network edge. The opportunity to combine these two domains of technologies to power interconnected devices with intelligence is likely to pave the way for a new wave of technology revolutions. Embracing this technology revolution, in this article, we present a novel computing vision named Deep Edge Intelligence (DEI). DEI employs Deep Learning, Artificial Intelligence, Cloud and Edge Computing, 5G/6G networks, Internet of Things, Microservices, etc. aiming to provision reliable and secure intelligence services to every person and organisation at any place with better user experience. The vision, system architecture, key layers and features of DEI are also detailed. Finally, we reveal the key enabling technologies and research challenges associated with it.

cs.DC↗

arXiv2021

CSSR: A Context-Aware Sequential Software Service Recommendation Model

We propose a novel software service recommendation model to help users find their suitable repositories in GitHub. Our model first designs a novel context-induced repository graph embedding method to leverage rich contextual information of repositories to alleviate the difficulties caused by the data sparsity issue. It then leverages sequence information of user-repository interactions for the first time in the software service recommendation field. Specifically, a deep-learning based sequential recommendation technique is adopted to capture the dynamics of user preferences. Comprehensive experiments have been conducted on a large dataset collected from GitHub against a list of existing methods. The results illustrate the superiority of our method in various aspects.

cs.IR↗

arXiv2021

Privacy-Aware Identity Cloning Detection based on Deep Forest

We propose a novel method to detect identity cloning of social-sensor cloud service providers to prevent the detrimental outcomes caused by identity deception. This approach leverages non-privacy-sensitive user profile data gathered from social networks and a powerful deep learning model to perform cloned identity detection. We evaluated the proposed method against the state-of-the-art identity cloning detection techniques and the other popular identity deception detection models atop a real-world dataset. The results show that our method significantly outperforms these techniques/models in terms of Precision and F1-score.

cs.SI↗

arXiv2021

NPS-AntiClone: Identity Cloning Detection based on Non-Privacy-Sensitive User Profile Data

Social sensing is a paradigm that allows crowdsourcing data from humans and devices. This sensed data (e.g. social network posts) can be hosted in social-sensor clouds (i.e. social networks) and delivered as social-sensor cloud services (SocSen services). These services can be identified by their providers' social network accounts. Attackers intrude social-sensor clouds by cloning SocSen service providers' user profiles to deceive social-sensor cloud users. We propose a novel unsupervised SocSen service provider identity cloning detection approach, NPS-AntiClone, to prevent the detrimental outcomes caused by such identity deception. This approach leverages non-privacy-sensitive user profile data gathered from social networks to perform cloned identity detection. It consists of three main components: 1) a multi-view account representation model, 2) an embedding learning model and 3) a prediction model. The multi-view account representation model forms three different views for a given identity, namely a post view, a network view and a profile attribute view. The embedding learning model learns a single embedding from the generated multi-view representation using Weighted Generalized Canonical Correlation Analysis. Finally, NPS-AntiClone calculates the cosine similarity between two accounts' embedding to predict whether these two accounts contain a cloned account and its victim. We evaluated our proposed approach using a real-world dataset. The results showed that NPS-AntiClone significantly outperforms the existing state-of-the-art identity cloning detection techniques and machine learning approaches.

cs.SI↗

arXiv2021

Conflict Detection in IoT-based Smart Homes

We propose a novel framework that detects conflicts in IoT-based smart homes. Conflicts may arise during interactions between the resident and IoT services in smart homes. We propose a generic knowledge graph to represent the relations between IoT services and environment entities. We also profile a generic knowledge graph to a specific smart home setting based on the context information. We propose a conflict taxonomy to capture different types of conflicts in a single resident smart home setting. A conflict detection algorithm is proposed to identify potential conflicts using the profiled knowledge graph. We conduct a set of experiments on real datasets and synthesized datasets to validate the effectiveness and efficiency of our proposed approach.

cs.AI↗

arXiv2021

Social Media Identity Deception Detection: A Survey

Social media have been growing rapidly and become essential elements of many people's lives. Meanwhile, social media have also come to be a popular source for identity deception. Many social media identity deception cases have arisen over the past few years. Recent studies have been conducted to prevent and detect identity deception. This survey analyses various identity deception attacks, which can be categorized into fake profile, identity theft and identity cloning. This survey provides a detailed review of social media identity deception detection techniques. It also identifies primary research challenges and issues in the existing detection techniques. This article is expected to benefit both researchers and social media providers.

cs.CR↗

arXiv2020

Subjective Metrics-based Cloud Market Performance Prediction

This paper explores an effective machine learning approach to predict cloud market performance for cloud consumers, providers and investors based on social media. We identified a set of comprehensive subjective metrics that may affect cloud market performance via literature survey. We used a popular sentiment analysis technique to process customer reviews collected from social media. Cloud market revenue growth was selected as an indicator of cloud market performance. We considered the revenue growth of Amazon Web Services as the stakeholder of our experiments. Three machine learning models were selected: linear regression, artificial neural network, and support vector machine. These models were compared with a time series prediction model. We found that the set of subjective metrics is able to improve the prediction performance for all the models. The support vector machine showed the best prediction results compared to the other models.

cs.SI↗

arXiv2020

Heuristics based Mosaic of Social-Sensor Services for Scene Reconstruction

We propose a heuristics-based social-sensor cloud service selection and composition model to reconstruct mosaic scenes. The proposed approach leverages crowdsourced social media images to create an image mosaic to reconstruct a scene at a designated location and an interval of time. The novel approach relies on the set of features defined on the bases of the image metadata to determine the relevance and composability of services. Novel heuristics are developed to filter out non-relevant services. Multiple machine learning strategies are employed to produce smooth service composition resulting in a mosaic of relevant images indexed by geolocation and time. The preliminary analytical results prove the feasibility of the proposed composition model.

cs.CV↗