Learning from learning machines: a new generation of AI technology to\n meet the needs of science

We outline emerging opportunities and challenges to enhance the utility of AI\nfor scientific discovery. The distinct goals of AI for industry versus the\ngoals of AI for science create tension between identifying patterns in data\nversus discovering patterns in the world from data. If we address the\nfundamental challenges associated with "bridging the gap" between domain-driven\nscientific models and data-driven AI learning machines, then we expect that\nthese AI models can transform hypothesis generation, scientific discovery, and\nthe scientific process itself.\n