The Science of Curiosity

01/04/2021 to 31/12/2025

We want to build a theory of child-like curiosity. Children are arguably the only known
system that demonstrably and reproducibly develops into intelligent agents through play-
ful exploration. We want to build machines that do the same. This involves studying how
children explore their environment during free play, extracting the algorithms they apply,
and using these models to build robots that effectively explore similar environments. Com-
putationally mirroring this development requires a formal understanding of curiosity – the
ability to explore environments in the absence of rewards. Studying curiosity demands an
inter-disciplinary approach, where developmental psychologists, cognitive scientists, and
roboticists work together to understand the human ability to be curious and build algo-
rithms that mirror this ability. Our proposed project will coalesce around three objectives.
In the first objective, we will study curiosity by letting children play freely while tracking their
actions. In the second objective, we will identify the best model to describe children’s be-
havior, by building cognitive algorithms of child-like curiosity. Finally, in the third objective,
we will build robots that can efficiently solve similar tasks. Our ultimate goal is to build more
powerful robots that play like children, thereby moving towards a science of curiosity.