Cognitive systems are starting to be deployed as appliances across the technological landscape of modern societies. The increasing availability of high performance computing platforms has opened an opportunity for statistics-based cognitive systems that per- form quite as humans in certain tasks that resisted the symbolic methods of classic artificial intelligence. Cognitive artefacts appear every day in the media, raising a wave of mild fear concerning artificial intelligence and its impact on society. These systems, performance notwithstanding, are quite brittle and their reduced dependability limits their potential for massive deployment in mission-critical applications —e.g. in autonomous driving or medical diagnosis. In this paper we explore the actual possibility of building cognitive systems using engineering-grade methods that can assure the satisfaction of strict requirements for their operation. The final conclusion will be that, besides the potential improvement provided by a rigorous engineering process, we are still in need of a solid theory —possibly the main outcome of cognitive science— that could sustain such endeavour. In this sense, we propose the use of formal ontologies as back- bones of cognitive systems engineering processes and workflows.

Ontologies as Backbone of Cognitive Systems Engineering. Ricardo Sanz, Julita Bermejo, Juan Morago and Carlos Hernández. AISB Symposium on Cognition And OntologieS (CAOS) 2017

Draft paper @ ASLab

A major challenge when building complex and critical systems is the management of change in the system and in its operational environment. The increasing complexity forces autonomous systems to detect critical changes to avoid their progress towards undesirable states. We need new methods to build systems that can tune their adaptability protocols, transferring the control of uncertainty to their inner domain to strive for wellness. In essence, these are mechanisms to impose the fulfillment of system-wide wellness requirements to reduce the influence of the outer domain to be fully driven by the influence of the inner one. From the stance of cognitive systems, biological emotion suggests a strategy to configure value-based systems to use semantic self-representations of the state. A method inspired by emotion theories can causally connect the inner domain of the system and its objectives of wellness, focusing on dynamically adapting the system to avoid the progress of critical states. This method shall endow the system with a transversal mechanism to monitor its inner processes, detecting critical states and managing its adaptivity in order to maintain the wellness goals. The paper describes the current vision produced by this work-in-progress.

Emotions and the engineering of adaptiveness in complex systems. M.G. sanchez-Escribano & R. Sanz. INCOSE Conference on Systems Engineering Research (CSER 2014)

Draft paper @ ASLab

We have designed a machine that becomes increasingly better at behaving in underspecified circumstances, in a goal-directed way, on the job, by modeling itself and its environment as experience accumulates. Based on principles of autocatalysis, endogeny, and reflectivity, the work provides an architectural blueprint for constructing systems with high levels of operational autonomy in underspecified circumstances, starting from a small seed. Through value-driven dynamic priority scheduling controlling the parallel execution of a vast number of reasoning threads, the system achieves recursive self-improvement after it leaves the lab, within the boundaries imposed by its designers. A prototype system has been implemented and demonstrated to learn a complex real-world task, real-time multimodal dialogue with humans, by on-line observation. Our work presents solutions to several challenges that must be solved for achieving artificial general intelligence.

Bounded Recursive Self-Improvement. E. Nivel, K. R. Thórisson, B. R. Steunebrink, H. Dindo, G. Pezzulo, M. Rodriguez, C. Hernandez, D. Ognibene, J. Schmidhuber, R. Sanz, H. P. Helgason, A. Chella, G. K. Jonsson

Article @ arXiv