How This Works (or soon will)
Train neural net to learn a mapping of environmental features to which algorithm performs best under those conditions
Then, during run time the evaluator monitors the environment and uses that neural net to determine which algorithm to have running
Swaps in new algorithm when necessary