Machine learning has become ubiquitous across many fields in the last decade and modern real world applications often require a decentralized solution for training such models. This demand sprouted the research in federated learning, which solves some of the challenges with centralized machine learning, but at the same times raises further questions in regard to security, privacy and scalability. We have designed and implemented DecFL, an ubiquitous protocol for decentralized model training. The protocol is machine-learning-model-, vendor-, and technology-agnostic and provides a basis for practitioner's own implementations. The implemented DecFL framework presented in this article is an exemplary realization of the carefully designed protocol stack based on Ethereum and IPFS and offers a scalable baseline solution for decentralized machine learning. In this article, we present a study based on the proposed protocol, its theoretical bounds and experiments based on the implemented framework. Using open-source datasets (MNIST and CIFAR10), we demonstrate key features, the actual cost of training a model (in euro) and the communication overhead. We further show that through a proper choice of technologies DecFL achieves a linear scaling, which is a non-trivial task in a decentralized setting. Along with discussing some of the security challenges in the field, we highlight aggregation poisoning as a relevant attack vector, its associated risks and a possible prevention strategy for decentralized model training through DecFL.
{"title":"DecFL: An Ubiquitous Decentralized Model Training Protocol and Framework Empowered by Blockchain","authors":"Felix Morsbach, S. Toor","doi":"10.1145/3457337.3457842","DOIUrl":"https://doi.org/10.1145/3457337.3457842","url":null,"abstract":"Machine learning has become ubiquitous across many fields in the last decade and modern real world applications often require a decentralized solution for training such models. This demand sprouted the research in federated learning, which solves some of the challenges with centralized machine learning, but at the same times raises further questions in regard to security, privacy and scalability. We have designed and implemented DecFL, an ubiquitous protocol for decentralized model training. The protocol is machine-learning-model-, vendor-, and technology-agnostic and provides a basis for practitioner's own implementations. The implemented DecFL framework presented in this article is an exemplary realization of the carefully designed protocol stack based on Ethereum and IPFS and offers a scalable baseline solution for decentralized machine learning. In this article, we present a study based on the proposed protocol, its theoretical bounds and experiments based on the implemented framework. Using open-source datasets (MNIST and CIFAR10), we demonstrate key features, the actual cost of training a model (in euro) and the communication overhead. We further show that through a proper choice of technologies DecFL achieves a linear scaling, which is a non-trivial task in a decentralized setting. Along with discussing some of the security challenges in the field, we highlight aggregation poisoning as a relevant attack vector, its associated risks and a possible prevention strategy for decentralized model training through DecFL.","PeriodicalId":270073,"journal":{"name":"Proceedings of the 3rd ACM International Symposium on Blockchain and Secure Critical Infrastructure","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130870002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Session details: BSCI Session 1","authors":"Jiamou Liu","doi":"10.1145/3467995","DOIUrl":"https://doi.org/10.1145/3467995","url":null,"abstract":"","PeriodicalId":270073,"journal":{"name":"Proceedings of the 3rd ACM International Symposium on Blockchain and Secure Critical Infrastructure","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130773145","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Vinayak Iyer, Kshitij Shah, S. Rane, R. Shankarmani
Agriculture is the lifeline of Indian economy which contributes to 22% of the GDP, provides 58% employment, and sustains 69% of the population. However, agricultural production comes with various risk factors beyond human control. Farmers are highly exposed to risk and it is aggravated by the limited span of risk mitigation instruments such as insurance where there is a lack of participation from the private sector and the claim payout procedure is very lengthy, due to which the farmer cannot initiate new cropping activity or it is delayed. There is no proper and adequate provision of security for crops grown by Indian farmers and despite increasing vulnerability, they have not yet fully developed the capacity to mitigate risk effectively. There is a need to design innovative insurance products to balance the interests of the insurer as well as the insured. With the Prime Minister's call to be "self sufficient" or "aatmanirbhar", in this paper we present a decentralised peer-to-peer crop insurance framework which enables to overcome the current disadvantages by securing farmers' interests, eliminating intermediaries and providing a secure, standardized, and transparent system that does not hinder any information from stakeholders. Through blockchain technology, the system promotes trust in a trust less environment wherein the farmers and the private investors go into a contract. The investors pool their investments and provide cover for the farmers, thus help the nation become self sufficient and help the farmers prosper.
{"title":"Decentralised Peer-to-Peer Crop Insurance","authors":"Vinayak Iyer, Kshitij Shah, S. Rane, R. Shankarmani","doi":"10.1145/3457337.3457837","DOIUrl":"https://doi.org/10.1145/3457337.3457837","url":null,"abstract":"Agriculture is the lifeline of Indian economy which contributes to 22% of the GDP, provides 58% employment, and sustains 69% of the population. However, agricultural production comes with various risk factors beyond human control. Farmers are highly exposed to risk and it is aggravated by the limited span of risk mitigation instruments such as insurance where there is a lack of participation from the private sector and the claim payout procedure is very lengthy, due to which the farmer cannot initiate new cropping activity or it is delayed. There is no proper and adequate provision of security for crops grown by Indian farmers and despite increasing vulnerability, they have not yet fully developed the capacity to mitigate risk effectively. There is a need to design innovative insurance products to balance the interests of the insurer as well as the insured. With the Prime Minister's call to be \"self sufficient\" or \"aatmanirbhar\", in this paper we present a decentralised peer-to-peer crop insurance framework which enables to overcome the current disadvantages by securing farmers' interests, eliminating intermediaries and providing a secure, standardized, and transparent system that does not hinder any information from stakeholders. Through blockchain technology, the system promotes trust in a trust less environment wherein the farmers and the private investors go into a contract. The investors pool their investments and provide cover for the farmers, thus help the nation become self sufficient and help the farmers prosper.","PeriodicalId":270073,"journal":{"name":"Proceedings of the 3rd ACM International Symposium on Blockchain and Secure Critical Infrastructure","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121628888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Session details: BSCI Short Paper Session 2","authors":"Xuyun Zhang","doi":"10.1145/3467998","DOIUrl":"https://doi.org/10.1145/3467998","url":null,"abstract":"","PeriodicalId":270073,"journal":{"name":"Proceedings of the 3rd ACM International Symposium on Blockchain and Secure Critical Infrastructure","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133066243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abhilash Kancharla, Jongho Seol, Hyeyoung Kim, N. Park
This paper proposes a quantitative model for a new blockchain technology that distributes public ledger in a decentralized manner, referred to as Distributed Decentralized chain (DDC). A clique of k nodes in the P2P network participates in storing a complete copy of blockchain instead of having every node carry an entire copy. The proposed model is k-Queue Variable Bulk Arrival and Static Bulk Service queueing model (k-VBASBS), in which the state is defined by (i,k), where i is the number of slots from 0 upto n in a block on the current node, and k is the number of distributed nodes to store a complete copy of a chain of blocks. Without loss of generality and practicality, it is assumed that there are two different transaction posting rates assumed to take into account the overhead of inter-node (i.e., μ(inter-node)) control-hopping versus the one of the original intra-node posting rate (i.e., μ(intra-node)), and μ(inter-node)μ<<(intra-node). Based on the proposed k-VBASBS model, the average waiting time, space requirement, and throughput of the transactions will be simulated for the performance, and the dependability will be also modeled and simulated by the vulnerability to 51% attack.
{"title":"Distributed Decentralized Chain (DDC) and k-Queue Variable Bulk Arrival and Static Bulk Service Model","authors":"Abhilash Kancharla, Jongho Seol, Hyeyoung Kim, N. Park","doi":"10.1145/3457337.3457840","DOIUrl":"https://doi.org/10.1145/3457337.3457840","url":null,"abstract":"This paper proposes a quantitative model for a new blockchain technology that distributes public ledger in a decentralized manner, referred to as Distributed Decentralized chain (DDC). A clique of k nodes in the P2P network participates in storing a complete copy of blockchain instead of having every node carry an entire copy. The proposed model is k-Queue Variable Bulk Arrival and Static Bulk Service queueing model (k-VBASBS), in which the state is defined by (i,k), where i is the number of slots from 0 upto n in a block on the current node, and k is the number of distributed nodes to store a complete copy of a chain of blocks. Without loss of generality and practicality, it is assumed that there are two different transaction posting rates assumed to take into account the overhead of inter-node (i.e., μ(inter-node)) control-hopping versus the one of the original intra-node posting rate (i.e., μ(intra-node)), and μ(inter-node)μ<<(intra-node). Based on the proposed k-VBASBS model, the average waiting time, space requirement, and throughput of the transactions will be simulated for the performance, and the dependability will be also modeled and simulated by the vulnerability to 51% attack.","PeriodicalId":270073,"journal":{"name":"Proceedings of the 3rd ACM International Symposium on Blockchain and Secure Critical Infrastructure","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124262817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Canh Tran, Van-Duy Pham, Thang Nguyen, Huu Dinh, Minh-Tri Hoang, Thanh-Chung Dao, B. Nguyen, Ba-Lam Do
In recent years, the proliferation of blockchain technology has opened many research directions. In this context, the combination of blockchain-based techniques and traditional methods to improve existing systems has received significant interest from researchers. In this paper, we present a decentralized storage and sharing system based on a combination of IPFS (Inter-Planetary File System), encryption technologies (including Advanced Encryption Standard (AES), Elliptic Curve Cryptosystem (ECC), ABE (Attribute-based Encryption), Multi-Authority ABE (MA-ABE)), and multichain. In particular, we facilitate the advantages of the IPFS network to store user's data in a distributed manner. Furthermore, we make use of a cryptographer to protect the privacy of data. The hash returned by the IPFS network will be stored in our multichain architecture to provide transparency for all users participating in the system. To the best of our knowledge, it is the first storage and sharing system using IPFS, cryptographer, and multichain to ensure decentralized, trustworthy, transparent characteristics for storing and sharing data.
{"title":"A Novel Approach for Developing Decentralized Storage and Sharing Systems","authors":"Canh Tran, Van-Duy Pham, Thang Nguyen, Huu Dinh, Minh-Tri Hoang, Thanh-Chung Dao, B. Nguyen, Ba-Lam Do","doi":"10.1145/3457337.3457845","DOIUrl":"https://doi.org/10.1145/3457337.3457845","url":null,"abstract":"In recent years, the proliferation of blockchain technology has opened many research directions. In this context, the combination of blockchain-based techniques and traditional methods to improve existing systems has received significant interest from researchers. In this paper, we present a decentralized storage and sharing system based on a combination of IPFS (Inter-Planetary File System), encryption technologies (including Advanced Encryption Standard (AES), Elliptic Curve Cryptosystem (ECC), ABE (Attribute-based Encryption), Multi-Authority ABE (MA-ABE)), and multichain. In particular, we facilitate the advantages of the IPFS network to store user's data in a distributed manner. Furthermore, we make use of a cryptographer to protect the privacy of data. The hash returned by the IPFS network will be stored in our multichain architecture to provide transparency for all users participating in the system. To the best of our knowledge, it is the first storage and sharing system using IPFS, cryptographer, and multichain to ensure decentralized, trustworthy, transparent characteristics for storing and sharing data.","PeriodicalId":270073,"journal":{"name":"Proceedings of the 3rd ACM International Symposium on Blockchain and Secure Critical Infrastructure","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124944144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yipeng Ji, Jingyi Wang, Shaoning Li, Yangyang Li, Shenwen Lin, Xiong Li
Anomaly event detection is crucial for critical infrastructure security(transportation system, social-ecological sector, insurance service, government sector etc.) due to its ability to reveal and address the potential cyber-threats in advance by analysing the data(messages, microblogs, logs etc.) from digital systems and networks. However, the convenience and applicability of smart devices and the maturity of connected technology make the social anomaly events data multi-source and dynamic, which result in the inadaptability for multi-source data detection and thus affect the critical infrastructure security. To effectively address the proposed problems, in this paper, we design a novel anomaly detection method based on multi-source data. First, we leverage spectral clustering algorithm for feature extraction and fusion of multiple data sources. Second, by harnessing the power of deep graph neural network(Deep-GNN), we perform a fine-gained anomaly social event detection, revealing the threatening events and guarantee the critical infrastructure security. Experimental results demonstrate that our framework outperforms other baseline anomaly event detection methods and shows high tracking accuracy, strong robustness and stability.
{"title":"An Anomaly Event Detection Method Based on GNN Algorithm for Multi-data Sources","authors":"Yipeng Ji, Jingyi Wang, Shaoning Li, Yangyang Li, Shenwen Lin, Xiong Li","doi":"10.1145/3457337.3457846","DOIUrl":"https://doi.org/10.1145/3457337.3457846","url":null,"abstract":"Anomaly event detection is crucial for critical infrastructure security(transportation system, social-ecological sector, insurance service, government sector etc.) due to its ability to reveal and address the potential cyber-threats in advance by analysing the data(messages, microblogs, logs etc.) from digital systems and networks. However, the convenience and applicability of smart devices and the maturity of connected technology make the social anomaly events data multi-source and dynamic, which result in the inadaptability for multi-source data detection and thus affect the critical infrastructure security. To effectively address the proposed problems, in this paper, we design a novel anomaly detection method based on multi-source data. First, we leverage spectral clustering algorithm for feature extraction and fusion of multiple data sources. Second, by harnessing the power of deep graph neural network(Deep-GNN), we perform a fine-gained anomaly social event detection, revealing the threatening events and guarantee the critical infrastructure security. Experimental results demonstrate that our framework outperforms other baseline anomaly event detection methods and shows high tracking accuracy, strong robustness and stability.","PeriodicalId":270073,"journal":{"name":"Proceedings of the 3rd ACM International Symposium on Blockchain and Secure Critical Infrastructure","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129606400","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nami Ashizawa, Naoto Yanai, Jason Paul Cruz, Shingo Okamura
Ethereum smart contracts are programs that run on the Ethereum blockchain, and many smart contract vulnerabilities have been discovered in the past decade. Many security analysis tools have been created to detect such vulnerabilities, but their performance decreases drastically when codes to be analyzed are being rewritten. In this paper, we propose Eth2Vec, a machine-learning-based static analysis tool for vulnerability detection in smart contracts. It is also robust against code rewrites, i.e., it can detect vulnerabilities even in rewritten codes. Existing machine-learning-based static analysis tools for vulnerability detection need features, which analysts create manually, as inputs. In contrast, Eth2Vec automatically learns features of vulnerable Ethereum Virtual Machine (EVM) bytecodes with tacit knowledge through a neural network for natural language processing. Therefore, Eth2Vec can detect vulnerabilities in smart contracts by comparing the code similarity between target EVM bytecodes and the EVM bytecodes it already learned. We conducted experiments with existing open databases, such as Etherscan, and our results show that Eth2Vec outperforms a recent model based on support vector machine in terms of well-known metrics, i.e., precision, recall, and F1-score.
{"title":"Eth2Vec: Learning Contract-Wide Code Representations for Vulnerability Detection on Ethereum Smart Contracts","authors":"Nami Ashizawa, Naoto Yanai, Jason Paul Cruz, Shingo Okamura","doi":"10.1145/3457337.3457841","DOIUrl":"https://doi.org/10.1145/3457337.3457841","url":null,"abstract":"Ethereum smart contracts are programs that run on the Ethereum blockchain, and many smart contract vulnerabilities have been discovered in the past decade. Many security analysis tools have been created to detect such vulnerabilities, but their performance decreases drastically when codes to be analyzed are being rewritten. In this paper, we propose Eth2Vec, a machine-learning-based static analysis tool for vulnerability detection in smart contracts. It is also robust against code rewrites, i.e., it can detect vulnerabilities even in rewritten codes. Existing machine-learning-based static analysis tools for vulnerability detection need features, which analysts create manually, as inputs. In contrast, Eth2Vec automatically learns features of vulnerable Ethereum Virtual Machine (EVM) bytecodes with tacit knowledge through a neural network for natural language processing. Therefore, Eth2Vec can detect vulnerabilities in smart contracts by comparing the code similarity between target EVM bytecodes and the EVM bytecodes it already learned. We conducted experiments with existing open databases, such as Etherscan, and our results show that Eth2Vec outperforms a recent model based on support vector machine in terms of well-known metrics, i.e., precision, recall, and F1-score.","PeriodicalId":270073,"journal":{"name":"Proceedings of the 3rd ACM International Symposium on Blockchain and Secure Critical Infrastructure","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134078724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：