Lecture: Dr. Keyan Ghazi-Zahedi (Leipzig): Quantifying Morphological Computation
NMK Special Lecture
Dr. Keyan Ghazi-Zahedi (Max Planck Institute for Mathematics in the Sciences, Leipzig, Information Theory of Cognitive Systems Group):
Quantifying Morphological Computation
In the context of embodied artificial intelligence, morphological computation originally refers to processes, which are conducted by the body (and environment) that otherwise would have to be performed by the brain. Exploiting environmental and morphological properties are an important feature of embodied systems. The main reason is that it allows to significantly reduce the controller complexity. An important aspect of morphological computation is that it cannot be assigned to an embodied system per se, but that it is, as I will show, behaviour and state dependent. The talk is divided into three parts. In the first part, I will discuss a quantification of morphological computation, which will then be applied to automated design of soft robots and muscle driven hopping models in the second part of the talk. In the final part, I will critically discuss the concept of morphological computation and present a possible unifying perspective on morphological contributions to a behaviour.
Host: Daniel Häufle
- Werner Reichardt Centre for Integrative Neuroscience
- Hertie Institute for Clinical Brain Research