OpenAI became the first to show human-like feats of dexterity with a robot hand in 2019, but it soon disbanded its robotics team.
Its set up had included a cage holding 19 cameras and more than 6,000 central processing units (CPUs) to learn huge neural networks which could control the hands.
Prof Lepora wanted to see if similar results could be achieved with more cost-efficient methods.
Four teams from the Massachusetts Institute of Technology, the University of California in Berkeley, Columbia University in New York, and Bristol achieved robotic hand dexterity feats using “simple set-ups” and desktop computers.