The robotic scientist

For centuries, scientists have attempted to identify and document analytical laws that underlie physical phenomena in nature. Despite the prevalence of computing power and data in scientific inquiry, finding natural laws and their corresponding equations has resisted automation. A key challenge to finding analytic relationships automatically is defining algorithmically what makes a correlation in observed data significant and non-trivial. This challenge is becoming a critical bottleneck as we try to understand and control increasingly complex systems, from biology to engineering. Results will be shown for a variety of mechanical and biological systems, from robotics to metabolic networks.