The surge in interest and application of large language models (LLMs) has sparked a drive to fine-tune these models to suit specific applications, such as finance and medical science. However, concerns regarding data privacy have emerged, especially when multiple stakeholders aim to collaboratively enhance LLMs using sensitive data. In this scenario, federated learning becomes a natural choice, allowing decentralized fine-tuning without exposing raw data to central servers. Motivated by this, we investigate how data privacy can be ensured in LLM fine-tuning through practical federated learning approaches, enabling secure contributions from multiple parties to enhance LLMs. Yet, challenges arise: (1) despite avoiding raw data exposure, there is a risk of inferring sensitive information from model outputs, and (2) federated learning for LLMs incurs notable communication overhead. To address these challenges, this article introduces DP-LoRA, a novel federated learning algorithm tailored for LLMs. DP-LoRA preserves data privacy by employing a Gaussian mechanism that adds noise in weight updates, maintaining individual data privacy while facilitating collaborative model training. Moreover, DP-LoRA optimizes communication efficiency via low-rank adaptation, minimizing the transmission of updated weights during distributed training. The experimental results across medical, financial, and general datasets using various LLMs demonstrate that DP-LoRA effectively ensures strict privacy constraints while minimizing communication overhead.
Open access
2 source records
Privacy-Preserving Technologies in Data
Artificial Intelligence in Healthcare and Education
Spatial crowdsourcing (SC) has become a fundamental and emerging technology in Metaverse, facilitating the creation of immersive experiences through location-based services. In these systems, a central SC server leverages SC workers who physically travel to task locations to gather spatiotemporal environment data. However, conventional SC systems face two significant challenges: (1) the SC server, functioning as a centralized authority, can sometimes be unreliable, either due to intentional or unintentional misconduct, (2) to ensure efficient task assignment and validation, the location privacy of tasks and workers is openly accessible. In this study, we formally define location privacy preserved proof generation and verification problem (LP-PGVP) within an SC task matching scenario, with the aim to the above two challenges. Our proposed solution is a blockchain-based SC system (BlockSC), which provides a decentralized platform for task requesters and workers in the Metaverse context through calling smart contracts. We also introduce a ciphertext-based task matching scheme where task location access is granted only to eligible workers executing a task, benefiting from the design of geographic coordinate transformation and bilinear mapping methodology. To further demonstrate the task matching scheme’s operation and impact, we present an easy-to-understand case study. Our evaluation findings confirm that the proposed system effectively maintains location privacy for both SC workers and task requesters, without a considerable sacrifice in task matching efficiency.
In Federated Learning (FL), blockchain has been extensively used to achieve distributed and tamper-resistant data processing. However, typical Blockchain-based Federated Learning (BFL) rarely considers clients’ resource and computing limits. High-capacity clients may be sacrificed when all clients train on the same neural network. This paper proposes a Blockchain-based Heterogeneous Federated Learning (BlocFL) model to address the challenges above. BlocFL replaces the central server with a consortium blockchain, and several neural networks are employed for local training. Considering the challenges in resource allocation in BFL, especially in heterogeneous networks, we propose a consortium blockchain-based heterogeneous federated learning client selection method. The proposed method optimizes the choice of client nodes under the limits of computational resources. Experiment results demonstrate that our method can allocate appropriate neural network models to each client and effectively improve the efficiency of local training in HFL. It also can achieve a comparable level of accuracy to the baseline approach with similar training parameters.
Mobile crowdsourcing (MCS) is an emerging paradigm that enables the outsourcing of a complex task to a group of mobile devices. The ability to utilize the collective power of mobile devices and human intelligence makes MCS a significant tool in various scenarios. Nevertheless, it faces practical challenge in protecting user privacy due to the sensitive nature of information collected by mobile devices. Additionally, the inherent openness of MSC and the heterogeneity of mobile devices raise reliability concerns among participants. To address these challenges, by integrating Federated Learning with the pairwise additive masking technique and the Chinese Remainder Theorem, we propose a Blockchain-based Privacy-preserving Federated Learning (BPFL) framework for mobile crowdsourcing, which allows mobile participants to collaboratively solve a crowdsourced machine learning task while preserving privacy. Besides, it employs blockchain technology to record the training process in a transparent and tamper-proof ledger. This ledger guarantees the verifiability of aggregation results and the fair distribution of training rewards, thereby enhancing trust and fairness. We prove that our BPFL supports privacy protection and trust mechanism simultaneously and resists inference and collusion attacks. Experimental results show that our BPFL can achieve high performance in terms of computation cost, communication cost and model accuracy, which is friendly for mobile users with resource-constrained devices in MCS ecosystems.
In recent years, mobile crowdsensing has emerged as a key technology for digital city construction, which acquires sensing data and mining data values with low cost and high flexibility. However, limited by the centrality of the sensing system, the complexity of the sensing process, and the uncertainty of sensing users, mobile crowdsensing has been severely suffering downgrading problems of security reliability and quality of service (QoS). In this study, a two-stage game involving blockchain technique is proposed to deal with security and QoS problems in mobile crowdsensing. First, a blockchain-based mobile crowdsensing system with a dual currency mechanism is established in a distributed manner to ensure the reliability of the sensing system. Second, a consensus protocol is designed to strengthen the security of sensing data based on a credit-based verifiers declaration game. Third, a quality rating mechanism based on the workers attitude game is established to improve the service quality of sensing tasks. Finally, some numerical results demonstrate the secure and efficient performances of mobile crowdsensing with the proposed framework.
To improve the security of crowdsourcing, existing studies introduce blockchain to ensure reliability and utilize cryptography (e.g., encryption and zero-knowledge proof) to protect data privacy. Nevertheless, the crowdsourcing process may involve sensitive identity information, and identity protection remains unresolved during stages such as data submission and correct payment. Especially, when workers invoke a smart contract to submit data, it inevitably exposes their identities. Identity disclosure significantly impacts the credibility of crowdsourcing platforms. Existing solutions suggest solving the problem through anonymous token contracts such as Zether. However, tokens can easily result in fund freezing or extra information leakage. Moreover, invoking the contract to submit data will still disclose workers' blockchain accounts. To tackle the identity protection issue, in this paper, we propose AcCrowd which achieves worker anonymity and payment correctness in crowdsourcing systems atop blockchain. We first design a verifiable proxy submission mechanism for data submission, enabling workers to invoke contracts without disclosing their accounts. Then, we introduce and improve the BlockMaze architecture to replace previous anonymous token-based methods, enhancing privacy and flexibility. Besides, we designed a revealed payment mechanism that utilizes an adaptor signature to bind data reveal and reward payment together, simultaneously protecting the requester and worker. Our security and performance evaluations demonstrate the security and practicability of AcCrowd.
Xiaohu Du, Xiao Chen, Jialun Cao, Ming Wen · 6 authors
Federated learning (FL) is an emerging machine learning paradigm that aims to address the problem of isolated data islands. To preserve privacy, FL allows machine learning models and deep neural networks to be trained from decentralized data kept privately at individual devices. FL has been increasingly adopted in missioncritical fields such as finance and healthcare. However, bugs in FL systems are inevitable and may result in catastrophic consequences such as financial loss, inappropriate medical decision, and violation of data privacy ordinance. While many recent studies were conducted to understand the bugs in machine learning systems, there is no existing study to characterize the bugs arising from the unique nature of FL systems. To fill the gap, we collected 395 real bugs from six popular FL frameworks (Tensorflow Federated, PySyft, FATE, Flower, PaddleFL, and Fedlearner) in GitHub and StackOverflow, and then manually analyzed their symptoms and impacts, prone stages, root causes, and fix strategies. Furthermore, we report a series of findings and actionable implications that can potentially facilitate the detection of FL bugs.
Antonio Carovilla, Remo Pareschi, Francesco Salzano
Swarm robotics is a field that studies the design and coordination of large groups of robots that can perform complex tasks through collective behaviors. One of the challenges in swarm robotics is to find the optimal balance between decentralization and centralization, as each approach has its advantages and disadvantages. In this paper, we propose a semi-centralized framework that integrates blockchain technology for enhanced coordination and security in swarm robotic systems. Our framework uses a centralized control unit that serves as a coordination hub for the robotic swarm while also leveraging blockchain technology to provide a secure and distributed ledger for data storage and communication. Thus, blockchain technology mitigates the reduced resilience, which is the price to pay for introducing semi-centralized control in robotic swarms, by ensuring the distributed management of data and programs. In the paper, we exemplify the application of our semi-centralized framework to optimize a path-finding problem for a swarm of robots. We also characterize a series of levels in allocating the blockchain infrastructure to cope with the corresponding levels of security threats.
The metaverse, as a rapidly evolving socio-technical phenomenon, exhibits significant potential across diverse domains by leveraging Web3 (a.k.a. Web 3.0) technologies such as blockchain, smart contracts, and non-fungible tokens (NFTs). This survey aims to provide a comprehensive overview of the Web3 metaverse from a human-centered perspective. We (i) systematically review the development of the metaverse over the past 30 years, highlighting the balanced contributions from its core components: Web3, immersive convergence, and crowd intelligence communities, (ii) define the metaverse that integrates the Web3 community as the Web3 metaverse and propose an analysis framework from the community, society, and human layers to describe the features, missions, and relationships for each community and their overlapping sections, (iii) survey the state-of-the-art of the Web3 metaverse from a human-centered perspective, namely, the identity, field, and behavior aspects, and (iv) provide supplementary technical reviews. To the best of our knowledge, this work represents the first systematic, interdisciplinary survey on the Web3 metaverse. Specifically, we commence by discussing the potential for establishing decentralized identities (DID) utilizing mechanisms such as profile picture (PFP) NFTs, domain name NFTs, and soulbound tokens (SBTs). Subsequently, we examine land, utility, and equipment NFTs within the Web3 metaverse, highlighting interoperable and full on-chain solutions for existing centralization challenges. Lastly, we spotlight current research and practices about individual, intra-group, and inter-group behaviors within the Web3 metaverse, such as Creative Commons Zero license (CC0) NFTs, decentralized education, decentralized science (DeSci), and decentralized autonomous organizations (DAO). Furthermore, we share our insights into several promising directions, encompassing three key socio-technical facets of Web3 metaverse development.
Bo Gu, Weiwei Hu, Shimin Gong, Zhou Su · 5 authors
Mobile crowdsensing (MCS) has emerged as a new sensing paradigm that relies on the sensing capabilities of the crowd to aggregate data. Unlike traditional MCS systems, where sensing data are traded via a third-party sensing platform, we propose a distributed data trading framework and investigate the potential of consortium blockchain to ensure the privacy and security of data transactions in MCS systems. The interactions between selling mobile users (SMUs) and buying mobile users (BMUs) are modeled as a Stackelberg game. Then, the amount of sensing time to purchase from each SMU and the price per unit sensing time are determined according to two auto-executing smart contracts. Notably, SMUs are compensated according to not only the amount of sensing time but also their reputation so that SMUs are encouraged to contribute high-quality data. Furthermore, the distributed ledger technology guarantees that the reputations of SMUs are updated and recorded in an immutable and traceable manner. Experimental results confirm that the proposed mechanism achieves near-optimal social welfare without requiring SMUs to know the price and data quality of each other.
This paper centers around a simple yet crucial question for everyday users: How should one choose their delegated validators within proof-of-stake (PoS) protocols, particularly in the context of Ethereum 2.0? This has been a long-overlooked gap, as existing studies have primarily focused on inter-committee (validator set) behaviors and activities, while neglecting the dynamic formation of committees, especially for individual stakeholders seeking reliable validators. Our study bridges this gap by diving into the delegation process (normal users delegate their small-value tokens to delegatees who later act as validators) before entering an actual consensus phase. We propose a Bayesian model to quantify normal users' trust in delegatees, which we further incorporate into a game-theoretical model to simulate users' reactions against a set of critical factors identified through extensive research (including 10+ staking service provider as well as 30+ PoS blockchains). Our results reveal that users tend to choose their delegatees and utilize their tokens by carefully weighing the delegation cost, the behaviors of other users, and the reputation of delegatees, ultimately reaching a Nash equilibrium. Unfortunately, the collective trend significantly increases the likelihood of token concentration on a small number of delegatees.
Eric Keilty, Keerthi Nelaturu, Anastasia Kastania, Andreas Veneris
The Move language provides superior security and verifiability compared to existing smart contract languages such as Solidity. As the language becomes more popular, gas optimization will become an important field of research. This paper presents the first work on gas optimization in the Move language. We chose Aptos as the underlying platform for our analysis since it is the leading Move-enabled blockchain platform, and it was the first to develop a gas meter. In this paper, we describe Aptos' gas meter in detail. Then, we analyze the vast research on gas optimization in Solidity, and explore how it can be implemented in the Move language. Finally, this paper proposes 11 gas optimization patterns and principles for the Move language, presents 5 patterns that decrease the time complexity of the smart contract but have no effect on gas consumption, and implements a sample smart contract for each proposed gas optimization pattern. Our results show that the proposed patterns reduce gas consumption in a typical smart contract by 7 – 56%.
Prasanna Siddharth Mukkamala, Haiqin Wu, Boris Düdder
Truth discovery is an effective and compelling approach to addressing data conflicts among different workers and offers more trustworthy truths to task requesters in crowdsourcing. Prior research either focused on studying more accurate truth discovery algorithms or aimed to protect data privacy from the centralized and honest-but-curious crowdsourcing platforms. They all overlooked the stronger threats from the malicious crowdsourcing platform (e.g., may return incorrectly estimated truths) and many critical issues inherited from centralization. This paper proposes a blockchain-based decentralized truth discovery scheme for crowdsourcing, with computation integrity guarantees against malicious participants and support for efficient processing of generic streaming data. We adopt the idea of hybrid storage and computations to ease the expensive on-chain cost. Workers are grouped for off-chain partial truth estimation and smart contracts are leveraged for on-chain final truth aggregation. To prevent any improper computations from malicious entities, we record the hashes of data and worker weights on-chain occasionally. Through theoretical analysis and extensive experiments over real-world and synthetic datasets implemented in Ethereum, we demonstrate that our scheme 1) achieves our reliability goals with certain privacy assurance; 2) exhibits a higher truth estimation accuracy than existing approaches and a lower gas consumption than the baseline.
Bogdan-Ionut Pahontu, Adrian Petcu, Alexandru Predescu, Diana Arsene · 5 authors
Water is one of the most important resources in our lives, and because of this, the interest in water management systems is growing constantly. A primary concern regarding urban water distribution is how to build robust solutions to facilitate water monitoring flows with the support of consumer involvement. Crowdsensing solutions contribute to the involvement in social platforms for increased awareness about the importance of water resources based on incentives and rewards. Blockchain is one of the technologies that has become increasingly popular in the last few years. The possibility of using this architecture in such different sectors while integrating emerging concepts, such as crowdsensing, the Internet of Things, serious gaming, and decision support systems, offers a lot of alternatives and approaches for designing modern applications. This paper aims to present how these technologies can be combined in order to migrate the functionalities of a water distribution management system from a centralized architecture to a decentralized one by leveraging blockchain technologies. The proposed application was designed to facilitate incident reporting flows in public water distribution networks. The proposed solution was to migrate the rewarding mechanisms using the Ethereum infrastructure. The novelty of this solution is determined by the introduction of this decentralized approach into the architecture and also by increasing customer interest by offering tradeable rewards and dynamic subscription discounts. This results in a new decentralized architecture that allows for more transparent interactions between the water provider and clients and increases customer engagement to contribute to water reporting flows.
Francisco Moya, Francisco J. Quesada, Luis Martı́nez, Francisco J. Estrella
Abstract Nowadays, Internet of Things (IoT) devices, especially wearable devices, are commonly integrated into modern intelligent healthcare software. These devices enable medical practitioners to monitor pervasively patients’ parameters outside the clinical environment. However, the ease of manipulating wearable devices and their data streams raises concerns regarding patient privacy and data trust. Distributed ledger technologies (DLT) offer solutions to enhance resistance against information manipulation and eliminate single points of failure. By leaveraging DLT, wearable-based solutions can be developed with a wider range of capabilities. This paper carries out an analysis of shortcomings, limitations, potential applications and needs in the medical domain, to introduce Phonendo 1.0, a DLT–IoT-based platform designed to capture data streams from wearable devices and publishing them on a distributed ledger technology infrastructure. The architecture and its difference services are justified based on the identified needs and challenges in the medical domain.
Crowdsourcing has emerged as a prevalent method for mitigating the risks of correctness and security in outsourced cloud computing. This process involves an aggregator distributing tasks, collecting responses, and aggregating outcomes from multiple data sources. Such an approach harnesses the wisdom of crowds to accomplish complex tasks, enhancing the accuracy of task completion while diminishing the risks associated with the malicious actions of any single entity. However, a critical question arises: How can we ensure that the aggregator performs its role honestly and each contributor's input is fairly evaluated? In response to this challenge, we introduce a novel protocol termed $\mathsf{zkTI}. This scheme guarantees both the honest execution of the aggregation process by the aggregator and the fair evaluation of each data source. It innovatively integrates a cryptographic construct known as zero-knowledge proof with a category of truth inference algorithms for the first time. Under this protocol, the aggregation operates with both correctness and verifiability, while ensuring fair assessment of data source reliability. Experimental results demonstrate the protocol's efficiency and robustness, making it a viable and effective solution in crowdsourcing and cloud computing.
Riccardo Trevisan, Mario Mureddu, Emilio Ghiani, Marco Galici · 5 authors
In the near future, renewable energy communities will play a crucial role in the transition to a cleaner energy system and the reduction of carbon emissions in the electricity sector. In order to ensure a broad base of participation among consumer and producers (prosumers), it is necessary to develop new business models and governance tools for energy communities, and, in this context, blockchain technology may have several applications in the implementation and management of these communities. This paper proposes a decentralized autonomous organization model for renewable energy communities, which will enable a transactive energy scenario for the community governance process. In order to illustrate the usefulness of the methodology, an example of the proposed approach with a distributed organization process for an automated local energy market among community members is presented.