We present a novel data-driven agent-based modeling framework to study innovation diffusion. Our first step is to learn a model of individual agent behavior from individual adoption characteristics. We then construct an agent-based simulation with the learned model embedded in artificial agents, and proceed to validate it using a holdout sequence of collective adoption decisions. Finally, we exemplify the proposed method can be used to explore and analyze a broad class of policies aimed at spurring innovation adoption. Keywords Machine Learning; Agent-based Modeling; Innovation Diffusion; Policy Optimization