How do people actively learn functional rules, i.e. a mapping of continuous inputs onto a continuous output? We investigate information search behavior in a multiple-feature function learning task in which participants either actively select or passively receive observations. We find that participants benefit from actively selecting information, in particular in their function extrapolation performance. By introducing and comparing different models of active function learning, we find that participants are best described by a non-parametric function learning model that learns about both the underlying function and inputs that are likely to produce high outputs. These results enrich our understanding of active function learning in complex domains.