Creating Agency and Cognition in Automated Systems: What can we learn from the Octopus? International Hybrid Workshop

In an attempt to create agency and cognition in automated systems, researchers have often drawn inspiration from the way that we humans act and think. This has led to the aim in Artificial Intelligence to create one agent that can perform a range of cognitive tasks by itself. Even with great successes in creating intelligent systems in AI, this latter aim has not been achieved yet. In this workshop we propose a different creature as an inspirational source for the creation of agency and cognition in automated systems: the octopus. Octopuses are intelligent creatures that seem to have multiple and layered cognitive systems that can both act in a uniform way as well as independently from one another. In Robotics, a layered architecture like that of Brooks’ creatures has already been shown to work well for robotic movement. Might it not be the case that this same kind of layering, but then in cognition, functions just as well for automated systems? This workshop will bring together a number of speakers from fields like biology, philosophy, AI and robotics, who will each take the octopus as a source of inspiration to look again at topics such as (artificial) cognition and agency.

Organization: Department of Philosophy, University of Konstanz Department of Collective Behavior, Max Planck Institute Konstanz Institute of Theoretical Physics, University of Innsbruck Department of Computer Science, Intelligent and Interactive Systems, University of Innsbruck.


The Program


1.) Eduardo Sampaio - “Biological features of the octopus: from neural architecture to multispecific hunting”

2.) Sidney Carls-Diamante - “The Octopus: implications for cognitive science”

3.) Erwan Renaudo - “A brief tour of robot's ‘cognition’ in autonomous robots”

4.) Maud van Lier - “Entangling embodiment and collaboration: a new approach to artificial agency”


Funding
Support by VolkswagenStiftung grant Az:97721 is gratefully acknowledged.