Remote connected science, hybrid human-machine learning in quantum physics
Prof. Dr. Jacob Friis Sherson (Department of Physics and Astronomy, Aarhus University, Denmark)

Spurred on by rapidly progressing scientific advances and massive investments, quantum technology is finally approaching the time to step out of university labs and into the corporate world. Real-world applications can be realized in two ways: a) by miniaturizing and mass producing quantum technology hardware or b) by creating centralized facilities on which it can be conducted assisted by a globally assessable interface. One example of the latter approach is the IBM quantum experience, which gives access to their quantum computing chip and has ushered in a time when theoreticians can experimentally test and develop their error correction models directly. An open question for the future of research is how to design interfaces that allow for an optimal interaction between human intuition, complex machinery, and the increasingly powerful machine learning algorithms.
In the www.scienceathome.org project, we have recently developed a gamified version of quantum optimization that has so far allowed 250,000 players from around the world to contribute to the research by providing insightful seeds for state-of-the-art optimization algorithms [1]. In recent work [2], we have launched our new open-access quantum research lab: an easily accessible remote interface for our ultra-cold atoms experiment allowing amateur scientists, students, and research institutions world-wide to perform state-of-the-art quantum experimentation. In the first test, a team of theoretical optimal control researchers employ a Remote version of their dCRAB optimization algorithm (RedCRAB), and secondly a gamified “democratic-lab” interface allowed 600 citizen scientists from around the world to participate in the optimization. In both cases solutions improving previous best performance were found. Within the new www.quatomic.org project, we are attempting to create visual software tools to enable intuition-based theoretical quantum research as well as a non-formalistic but powerful introduction to university level quantum education.
Finally, with a palette of additional games within cognitive science, behavioral economics, and corporate innovation we investigate the general features of individual and collaborative problem solving to shed additional light on the process of human intuition and innovation and potentially develop novel models of artificial intelligence.
[1] JJ Sørensen et al, Nature, 532, 210 (2016) [2] R. Heck et al, arXiv:1709.02230