Incremental Learning Implementations and Vision for Cyber Risk Detection in IoT

Abstract

The Internet of Things (IoT) has emerged as an Internet-based extension and considerably changed our world. A tremendous amount of IoT applications has highly eased people's daily lives, and improved allocations and usage of a broad range of resources (power bank sharing, bike sharing, etc.). However, the ingrained openness of underlying wireless systems renders these IoT entities vulnerable to a broad range of cyber risks, such as vulnerabilities of spectrum that can be a source of adversarial inference. Furthermore, as our reliance on wireless/connected devices and radios has also been dramatically increasing, public safety, business operations, socializing, and navigation as well as critical national communication infrastructures have become more vulnerable to cyber threats. In addition, the stream of data from IoT devices are also prone to cyber risks, which is of major concern for end users, and could also mislead users and ML models with wrong information during the training and learning phases. Hence, policymakers and industries have recently begun to perceive that the expansion of connected devices and their cyber susceptibility results in a large malicious and inferential risk. As a result, there is a need to identify and comprehend these risks in order to elaborate efficient security solutions. Moreover, the economic impact of IoT devices and associated security vulnerabilities are further growing in accordance with artificial intelligence (AI) integration into human-computer interaction (HCI), including banking, insurance, and so on. Consequently, cyber risks are growing in terms of both frequency and acuteness. Therefore, this article discusses the evolution of common real-time machine learning (RTML) approaches, which are considered as enablers for ensemble learning-based use cases. We discuss a cyber risk detection framework that can be built effectively using Boosting- and Bagging-based ensemble learning approaches.