Vedran Dunjko

Universität Innsbruck, Leiden University

Papers

Total Citations

H-Index

About

Vedran Dunjko is a leading researcher at the intersection of quantum computing and artificial intelligence, whose work is redefining how intelligent agents learn and generalize. His key research areas span quantum machine learning, reinforcement learning, and the development of autonomous robotic systems. Dunjko’s major contributions include pioneering the concept of projective simulation with generalization—a framework that enables artificial agents to not only learn from experience but also to extrapolate knowledge to novel situations, a critical capability for advanced AI. This work, published in 2017 and garnering 43 citations, established foundational criteria for generalization in learning environments where it is otherwise impossible. Beyond quantum-inspired models, Dunjko has made significant strides in autonomous robotics, developing methods for skill acquisition through active learning and exploratory behavior composition. His 2020 paper on robotic playing (21 citations) and his 2017 work on creativity in autonomous systems (15 citations) demonstrate how robots can autonomously extend their problem-solving abilities by creatively applying previously learned skills to new objects and scenarios. Dunjko’s research is notable for bridging theoretical rigor with practical robotic applications, positioning him as a visionary in creating machines that learn, generalize, and adapt with increasing autonomy.