On November 6, 2012, Dr. Peter Lewis will give a talk about "Engineering Self-Aware Computing Systems" in the context of the SFB 901 colloquium.
Abstract:
Novel computing systems are increasingly being composed of large numbers of heterogeneous components, each with potentially different goals or local perspectives, and connected in networks which change over time. Management of such systems quickly becomes infeasible for humans. As such, future computing systems should be able to achieve advanced levels of autonomous adaptive behaviour. In order for a system to effectively adapt itself in such a context, its ability to be self-aware becomes important.
There are several clusters of research in computer science and engineering which have used the term self-awareness explicitly. However, there is no general methodology or common framework for describing or benchmarking the self-awareness capabilities of these systems, or the benefits that self-awareness might bring. In this talk, I shall begin by surveying definitions and current understanding of self-awareness in psychology. I will then attempt to translate these concepts from psychology to the computing domain, and show how their explicit consideration may be beneficial in the engineering of adaptive computing systems. I will discuss how computational self-awareness may include knowledge of internal state, history, social or physical environment, goals, and perhaps even a system's own way of representing or reasoning about these things. Finally, I will describe our work in an example application domain, distributed smart-camera networks, where decentralised self-awareness can increase runtime adaptivity and robustness, and avoid the need for a priori information at design-time.
Biography:
Peter Lewis is a post-doctoral research fellow at the Centre of Excellence for Research in Computational Intelligence and Applications (Cercia) in the School of Computer Science at the University of Birmingham, UK. His current research is concerned with investigating algorithms and techniques for achieving self-awareness and self-expression in decentralized computational systems. Particular focuses include economics-inspired computational techniques and online learning algorithms, such as those using evolutionary computation and other nature-inspired techniques.