This chapter presents a project proposal, which defines future work in engineering the learning systems. This proposal outlines a number of directions in the fields of systems engineering, machine learning, knowledge engineering, and profile theory, that lead to the development of formal methods for the modeling and engineering of learning systems. This chapter describes a framework for formalisation and engineering the cognitive processes, which is based on applications of computational methods. The proposed work studies cognitive processes, and considers a cognitive system as a multi-agents system of human-cognitive agents. It is important to note that this framework can be applied to different types of learning systems, and there are various techniques from different theories (e.g., system theory, quantum theory, neural networks) can be used for the description of cognitive systems, which in turn can be represented by different types of cognitive agents.
Traditionally multi-agent learning is considered as the intersection of two subfields of artificial intelligence: multi-agent systems and machine learning. Conventional machine learning involves a single agent that is trying to maximize some utility function without any awareness of existence of other agents in the environment (Mitchell, 1997). Meanwhile, multi-agent systems consider mechanisms for the interaction of autonomous agents. Learning system is defined as a system where an agent learns to interact with other agents (e.g., Clouse, 1996; Crites & Barto, 1998; Parsons, Wooldridge & Amgoud, 2003). There are two problems that agents need to overcome in order to interact with each other to reach their individual or shared goals: since agents can be available/unavailable (i.e., they might appear and/or disappear at any time), they must be able to find each other, and they must be able to interact (Jennings, Sycara & Wooldridge, 1998).
Contemporary approaches to the modeling of learning systems in a multi-agent setting do not analyze nature of learning/cognitive tasks and quality of agents’ resources that have impact on the formation of multi-agent system and its learning performance. It is recognized that in most cognitively driven tasks, consideration of agents’ resource quality and their management may provide considerable improvement of performance process. However, most existing process models and conventional resource management approaches do not consider cognitive processes and agents’ resource quality (e.g., Norman, Preece, Chalmers, Jennings, Luck & Dang, 2003). Instead they overemphasize the technical components, resource existence/availability problems. For this reason, their practical utilisation is restricted to those applications where agents’ resources are not a critical variable. Formal representation and incorporation of cognitive processes in modeling frameworks is seen as very challenging for systems engineering research.
Therefore, future work in engineering the learning processes in cognitive system is considered with an emphasis on cognitive processes and knowledge/skills of cognitive agents as a resource in performance processes. There are many issues that need new and further research in engineering cognitive processes in learning system. New/novel directions in the fields of systems engineering, machine learning, knowledge engineering, and mathematical theories should be outlined to lead to the development of formal methods for the modeling and engineering of learning systems. This article describes a framework for formalisation and engineering the cognitive processes, which is based on applications of computational methods. The proposed work studies cognitive processes, and considers a cognitive system as a multi-agents system.
This project brings together work in systems engineering, knowledge engineering and machine learning for modelling cognitive systems and cognitive processes. A synthesis of formal methods and heuristic approaches to engineering tasks is used for the evaluation, comparison, analysis, evolution, and improvement of cognitive processes.