For centuries, scientists have attempted to identify analytical laws that underlie physical phenomena in nature. Despite today’s computing power, the process of finding natural laws and their corresponding equations has resisted automation. A key challenge to finding analytic relations automatically – that is, building an autonomous robot - is defining algorithmically what makes a correlation in observed data important and insightful. Scientists are gradually uncovering an ‘alphabet’ that can be used to describe the simplest to most complex physical systems – and by seeking dynamical invariants, researchers go from finding simple predictive models to discovering deeper natural laws. Dr. Lipson will demonstrate the process on a variety of mechanical, biological, and robotic systems.