Design and implementation of an adaptive learning system: An MSc project experience

Abstract

Individual project is a major component of MSc degrees in the UK in terms of allocated marks and overall weight. An MSc project typically lasts for up to 14-16 weeks, which makes it challenging to carry out elaborate research works. In this paper we highlight our experience in setting up one such project, which addresses the design and implementation of a bio-inspired adaptive learning system using cerebellar model articulation controller (CMAC) principles. To systematically execute the project, initially a guided and comprehensive literature survey was carried out to analyze in detail the CMAC design tradeoffs. Underpinning this analysis an adaptive design methodology was developed, leading to the implementation on a field programmble gate array (FPGA). The implementation was then validated using a real-time audio compression application. The validations demonstrated the advantages of adaptive learning systems in terms of reduced energy consumption with improved throughput and convergence. Moreover, it underlined the importance of a good plan and systematic delivery for a successful short-term research project.

Adaptive learning: a cluster-based literature review (2011-2022)

“…Li [104] proposes an adaptive online learning system that uses an improved ant colony algorithm to cater to learners' preferences, tastes, and knowledge levels. Martinez et al [113] focused on designing and implementing a bio-inspired adaptive learning system using cerebellar model articulation controller principles, which showed reduced energy consumption with improved throughput and convergence. Vo and Dutkiewicz [192] presents an algorithm for data segmentation in arrhythmia detection, utilizing an adaptive learning scheme to improve classification performance.…”