Single- versus Multiobjective Optimization for Evolution of Neural Controllers in Ms. Pac-Man

Table 2

Win rates for training results across all 10 runs.

Proposed algorithm	PAESNet_F	PAESNet_V	PAESNet_M
PAESNet_F	—	90%	90%
PAESNet_V	10%	—	60%
PAESNet_M	10%	40%	—

(𝐴 versus 𝐵) + (B versus A) = 100%.