Nonnative invasive species result in sizeable economic damages and expensive control costs. Because dynamic optimization models break down if controls depend in complex ways on past controls, non-uniform or scale-dependent spatial attributes, etc., decision support systems that allow learning may be preferred. We compare three models of an invasive weed in California’s grazing lands: (1) a stochastic dynamic programming model, (2) a reinforcement-based, experience-weighted attraction (EWA) learning model, and (3) an EWA model that also includes stochastic forage growth and penalties for repeated application of environmentally harmful control techniques. Results indicate that EWA learning models may be appropriate for invasive species management.