Each week, we publish layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.The article featured today is from the Canadian Journal of Statistics with the full article now available to read here.
Chen, F., Wan, H., Cai, H. and Cheng, G. (2021), Machine learning in/for blockchain: Future and challenges. Can J Statistics, 49: 1364-1382. https://doi.org/10.1002/cjs.11623
In recent years, machine learning and blockchain, the technology behind cryptocurrency (Bitcoin, Ethereum, Litecoin, etc.), have caught people’s attention. However, both technologies have their own disadvantages. For example, it is costly to label data and difficult to accelerate training process in machine learning, especially for deep learning. On the other hand, blockchain obtains and generates a huge amount of data, but cannot use them to improve the blockchain system itself.
Some recent research has demonstrated that combining the two technologies can complement each other’s weaknesses. As a result, we review existing works from the following two aspects. One is to review how machine learning is applied to the blockchain system for abnormal behavior detection, topological feature analysis, cryptocurrency price prediction, and bitcoin mining/validation strategy optimization. The other is to discuss how blockchain improves the security of the training data gathering and sharing in the collaborative learning system.