DescriptionTuning application parameters for optimal performance is a combinatorially challenging problem. Hence, techniques for modeling the functional relationships between various input features in the parameter space and application performance are important. We show that simple statistical inference techniques are inadequate to capture these relationships, and that even with more complex ensembles of models, the minimum coverage of the parameter space required via experimental observations is still quite large. We propose a deep learning-based approach that can combine the knowledge from exhaustive observations collected at a smaller scale with limited observations collected at a larger target scale. The proposed approach is able to accurately predict performance in the regimes of interest to performance analysts, while outperforming many traditional techniques. In particular, our approach can identify the best performing configurations when trained using as little as 1% of observations at the target scale.