GECCO'16 Model-Based Evolutionary Algorithms (MBEA) Workshop Chairs' Welcome

Fixed, problem-independent variation operators often fail to effectively exploit important features of highquality selected solutions, potentially leading to inefficient optimization in cases where a performance advantage can be gained by using variation operators that are informed by learnable problem features. One way to make variation operators more powerful and flexible is to  Model key features of solutions that influence their quality, and  generate a new population of candidate solutions using the model in the expectation of improved solution quality.