On the huge benefit of quasi-random mutations for multimodal optimization with application to grid-based tuning of neurocontrollers

In this paper, we study the optimization of a neural network used for controlling a Monte-Carlo Tree Search / Upper Confidence Trees (MCTS/UCT) algorithm. The main results are: (i) the specification of a new multimodal benchmark function; this function has been defined in particular in agreement with [1] which has pointed out that most multimodal functions are not satisfactory for some real-world multimodal scenarios (section 2); (ii) experimentation of Evolution Strategies on this new multimodal benchmark function, showing the great efficiency of quasi-random mutations in this framework (section 3); (iii) the proof-of-concept of the application of ES for grid-based tuning Neural Networks for controlling MCTS/UCT (see section 3).