Distributed Machine Learning Oriented Data Integrity Verification Scheme in Cloud Computing Environment

Abstract

Distributed Machine Learning (DML) is one of the core technologies for Artificial Intelligence (AI). However, in the existing distributed machine learning framework, the data integrity is not taken into account. If network attackers forge the data, modify the data, or destroy the data, the training model in the distributed machine learning system will be greatly affected, and the training results are led to be wrong. Therefore, it is crucial to guarantee the data integrity in the DML. In this paper, we propose a distributed machine learning oriented data integrity verification scheme (DML-DIV) to ensure the integrity of training data. Firstly, we adopt the idea of Provable Data Possession (PDP) sampling auditing algorithm to achieve data integrity verification so that our DML-DIV scheme can resist forgery attacks and tampering attacks. Secondly, we generate a random number, namely blinding factor, and apply the discrete logarithm problem (DLP) to construct proof and ensure privacy protection in the TPA verification process. Thirdly, we employ identity-based cryptography and two-step key generation technology to generate data owner's public/private key pair so that our DML-DIV scheme can solve the key escrow problem and reduce the cost of managing the certificates. Finally, formal theoretical analysis and experimental results show the security and efficiency of our DML-DIV scheme. INDEX TERMS Distributed machine learning, public auditing, data integrity, bilinear mapping, identity-based cryptography.

Distributed Data Integrity Verification Scheme in Multi-Cloud Environment

“…Unfortunately, this scheme did not support batch and distributed verification by multi-verifiers. The authors of [ 23 ] proposed a distributed machine learning-oriented data integrity verification scheme in a cloud computing environment. They adopted PDP as a sampling auditing algorithm in their scheme and generated a random number called a blinding factor and applied a discrete logarithm problem (DLP) to construct a proof and ensure privacy protection in the TPA verification process.…”

“…It removes the intermediary and enables peer-to-peer interactions between nodes, therefore enhancing trust. The other four works [ 17 , 20 , 21 , 23 ] did not provide these two points because they rely on the credibility of the third-party auditor (TPA), which is not ideal in real case circumstances. Distributed verification.…”

“…Thus, it can increase scalability and optimize time. The training data are stored in different storage locations with the nodes to reduce the burden [ 23 ]. So, external or internal threats exist that can endanger the integrity of training data and lead to the wrong training results.…”

Stub Signature-Based Efficient Public Data Auditing System using Dynamic Procedures in Cloud Computing

“…A comparison of three aspects of the suggested method with earlier research [4,7,11,12,[21][22][23] was later briefly described, as well as evidence of the suggested work's effectiveness. Figure 5 displays the computational cost of signature generation time with various data blocks 100,150,200,250,300,350,400 for the proposed scheme and the research papers [4,7,11,22,23].…”

Overmind: A Collaborative Decentralized Machine Learning Framework

“…Nowadays, a concept of distributed and decentralized artificial intelligence has been widely reported, which shows upcoming era in machine learning integrated fields of research such as artificial intelligence in distributed ledger technology [13], artificial intelligence as IoT [14,15], distributed security solution [16,17], etc. Surprisingly, these integrations are all based on a distributed and decentralized concept.…”