From distributed machine to distributed deep learning: a comprehensive survey

Abstract

Artificial intelligence has made remarkable progress in handling complex tasks, thanks to advances in hardware acceleration and machine learning algorithms. However, to acquire more accurate outcomes and solve more complex issues, algorithms should be trained with more data. Processing this huge amount of data could be time-consuming and require a great deal of computation. To address these issues, distributed machine learning has been proposed, which involves distributing the data and algorithm across several machines. There has been considerable effort put into developing distributed machine learning algorithms, and different methods have been proposed so far. We divide these algorithms in classification and clustering (traditional machine learning), deep learning and deep reinforcement learning groups. Distributed deep learning has gained more attention in recent years and most of the studies have focused on this approach. Therefore, we mostly concentrate on this category. Based on the investigation of the mentioned algorithms, we highlighted the limitations that should be addressed in future research.

DataMesh+: A Blockchain-Powered Peer-to-Peer Data Exchange Model for Self-Sovereign Data Marketplaces

“…Data mesh architecture [ 24 , 25 ], blockchain technology [ 26 , 27 , 28 ], and decentralized P2P storage networks [ 29 ] offer diverse features that P2P SSDMs can use to address the problems and limitations of existing centralized systems. Data mesh represents a novel approach to data architecture, transitioning from traditional centralized data management systems to a decentralized and domain-centric model [ 24 , 25 ].…”

“…Data mesh architecture [ 24 , 25 ], blockchain technology [ 26 , 27 , 28 ], and decentralized P2P storage networks [ 29 ] offer diverse features that P2P SSDMs can use to address the problems and limitations of existing centralized systems. Data mesh represents a novel approach to data architecture, transitioning from traditional centralized data management systems to a decentralized and domain-centric model [ 24 , 25 ]. This concept, pioneered by Zhamak Dehghani [ 24 ], revolves around treating data as a product and is based on four main pillars: (1) domain-oriented decentralized data ownership, (2) data as a product, (3) self-managed data infrastructure, and (4) federated computing management.…”

“…The emergence of data mesh architecture [ 24 , 25 ], blockchain [ 26 , 27 , 28 ], and decentralized P2P storage technologies [ 29 ] offers a paradigmatic solution to these pressing problems. Data mesh is a new conceptual data architecture framework that emphasizes decentralized ownership and control, treats data as a product, and focuses on domain-driven design [ 24 , 25 ].…”

“…The emergence of data mesh architecture [ 24 , 25 ], blockchain [ 26 , 27 , 28 ], and decentralized P2P storage technologies [ 29 ] offers a paradigmatic solution to these pressing problems. Data mesh is a new conceptual data architecture framework that emphasizes decentralized ownership and control, treats data as a product, and focuses on domain-driven design [ 24 , 25 ]. This approach enhances value extraction from data by overcoming the limitations of traditional centralized data systems across various business domains within or across large and complex organizations [ 25 , 30 ].…”

Refining the rheological characteristics of high drug loading ointment via SDS and machine learning

“…For the ointment preparation process, there are usually many parameters that need to be adjusted, such as temperature, pressure, reaction time, etc. Traditional methods may involve trial and error and experience, while machine learning technology can automate this process [ 13 , 14 ]. They learn the model by collecting and analyzing a large amount of data, and then optimize it according to the objective function.…”