Beneficial role of humans and AI in a machine learning age of the Telco ecosystem

Cassidy, Stephen; Perrett, Jay; Simons, Christopher

Abstract

The adoption of new technology in the telecommunications industry raises challenges, particularly when exploiting advances in artificial intelligence (AI) (e.g. dynamic optimisation and machine learning). Indeed, opinions have been expressed that AI will either replace everyone, gain sentience or not be able to contribute to anything meaningful where human creativity and innovation are required. Nonetheless, the need to exploit recent advances in AI in a dynamic telecommunications ecosystem of Internet of Things (IoT), 5G and rapidly diminishing margins should be unquestionable.In a recent panel of automation in Software Defined Networks (SDN) and Network Function Virtualisation (NFV), the CTO of a large European Operator cited the most significant problem to solve was not the use of AI in automation but rather its introduction alongside human operators. However, recent research in the area of interactive optimisation and machine learning at the University of the West of England, Bristol, UK, has incorporated humans "in-the-loop" with AI and found that superior solutions can be achieved by humans and AI working together rather than separately. To examine these issues, we present relevant case studies from BT Research Laboratories and Aria Networks / Facebook. We analyse the case study evidence and recent advances in AI research to carefully discern the crucial causative factors that underpin the situation. Based on the analysis, we conclude that the case for engineers working collaboratively "in-the-loop" with AI is compelling, and will be of great benefit to telecommunications businesses and the people involved in the areas of process automation and dynamic optimisation.

Citation

Cassidy, S., Perrett, J., & Simons, C. (2018, May). Beneficial role of humans and AI in a machine learning age of the Telco ecosystem. Paper presented at TM Forum Digital Transformation World, Nice, France

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: TM Forum Digital Transformation World
• Conference Location: Nice, France
• Start Date: May 14, 2018
• End Date: May 16, 2018
• Acceptance Date: May 15, 2018
• Publication Date: May 15, 2018
• Deposit Date: May 18, 2018
• Peer Reviewed: Not Peer Reviewed
• Keywords: telecommunications, interactive Artificial Intelligence
• Public URL: https://uwe-repository.worktribe.com/output/868074
• Additional Information: Title of Conference or Conference Proceedings : TM Forum Digital Transformation World