Inspired by challenges in designing energy technology policy in the face of climate change, we introduce an approach we call Robust Portfolio Decision Analysis, building on Belief Dominance as a prescriptive operationalization of a concept that has appeared in the literature under a number of names. The Belief Dominance concept synthesizes multiple conflicting sources of information to uncover alternatives that are intelligent responses in the presence of many beliefs or models. We use this concept to determine the set of non-dominated portfolios and to identify corresponding robust individual alternatives, thereby uncovering viable alternatives that may not be revealed otherwise. Using multiple large-scale expert elicitation studies and multiple Integrated Assessment Models, we illustrate how Robust Portfolio Decision Analysis helps identify robust R&D investments into individual technologies.