Singlet fission has shown potential for boosting the power conversion efficiency of solar cells, but the scarcity of suitable molecular materials hinders its implementation. We introduce an uncertainty-controlled genetic algorithm (ucGA) based on ensemble machine learning predictions from different molecular representations that concurrently optimizes excited state energies, synthesizability, and singlet exciton size for the discovery of singlet fission materials. We show that uncertainty in the model predictions can control how far the genetic optimization moves away from previously known molecules. Running the ucGA in an exploitative setup performs local optimization on variations of known singlet fission scaffolds, such as acenes. In an explorative mode, hitherto unknown candidates displaying excellent excited state properties for singlet fission are generated. We suggest a class of heteroatom-rich mesoionic compounds as acceptors for charge-transfer mediated singlet fission. When included in larger conjugated donor-acceptor systems, these units exhibit strong localization of the triplet state, favorable diradicaloid character and suitable triplet energies for exciton injection into semiconductor solar cells. As the proposed candidates are composed of fragments from synthesized molecules, they are likely synthetically accessible.