Practical optimal experiment design with probabilistic programs
This provides a general and principled method for scientists to automate experiment design, though it is incremental as it builds on existing probabilistic programming frameworks.
The paper tackles the problem of automating optimal experiment design for distinguishing competing theories by representing them as probabilistic programs, and demonstrates that the automatically designed experiments were empirically validated as optimal in cognitive psychology case studies.
Scientists often run experiments to distinguish competing theories. This requires patience, rigor, and ingenuity - there is often a large space of possible experiments one could run. But we need not comb this space by hand - if we represent our theories as formal models and explicitly declare the space of experiments, we can automate the search for good experiments, looking for those with high expected information gain. Here, we present a general and principled approach to experiment design based on probabilistic programming languages (PPLs). PPLs offer a clean separation between declaring problems and solving them, which means that the scientist can automate experiment design by simply declaring her model and experiment spaces in the PPL without having to worry about the details of calculating information gain. We demonstrate our system in two case studies drawn from cognitive psychology, where we use it to design optimal experiments in the domains of sequence prediction and categorization. We find strong empirical validation that our automatically designed experiments were indeed optimal. We conclude by discussing a number of interesting questions for future research.