Comprehensive overview on the deployment of machine learning, deep learning, reinforcement learning algorithms in Selfish mining attack in blockchain

Abstract

Blockchain, a disruptive technology, has many applications in the domain of Finance, banking, real estate, insurance, supply chain, gaming industry with much more plethora of applications in near future. In spite of the decentralized, distributed, transparent, tamper-proof, data- provenance nature of the blockchain, it is subject to a lot of security attacks such as forking attacks and block withholding attacks. One such attack under the forking attack is selfish mining, which targets the reward distribution and also the difficulty adjustment algorithms(DAA). An exhaustive surve is done on the existing approaches to detect selfish mining and also on the profitability of selfish mining attacks. This survey is organized particularly around the aspects of selection or exploration of the shortest branch of the blockchain when a fork occurs. We aim to identify the implications of selecting the shorter branch of the fork in the blockchain, especially after 2016 blocks, where a difficulty adjustment occurs. Our survey focuses on the deployment of machine learning and deep learning, reinforcement methods on mitigating the selfish mining attacks in the blockchain.