Your answer feels like its saying "its only partly random, you use the outcome of your experiments to walk the space of possible solutions more carefully, in the light of experience" which would make it more like GLMM or something markov-y