Physical Learning: Embodied Computing in Living and Non-Living Substrates

Catalog description

Adaptivity and learning are ubiquitous in living systems, from cytoplasm and biofilms to animal flocks to the origin of the complex nervous systems. Physical embodied learning where materials - living or non-living - learn and adapt remarkable range of behaviors only by exposure to physical environments is a fast evolving and exciting area in physics and biology. Going beyond the paradigm of traditional neuroscience, we will explore how can single cells learn and perform remarkable realtime behavior. Applying evolutionary principles, we will cover the evidence of gradual origin of the nervous system in metazoan tree of life. Laying broad foundations in computer science, we will explore range of examples of physical learning in mechanical, electrical, optical and fluidic systems. Finally, by highlighting the theoretical and experimental gaps in current understanding, we will explore these topics via class projects.