Abstract
The infeasible parts of the objective space in difficult many-objectiveoptimization problems cause trouble for evolutionary algorithms. This paperproposes a reference vector based algorithm which uses two interacting enginesto adapt the reference vectors and to evolve the population towards the truePareto Front (PF) s.t. the reference vectors are always evenly distributedwithin the current PF to provide appropriate guidance for selection. Thecurrent PF is tracked by maintaining an archive of undominated individuals, andadaptation of reference vectors is conducted with the help of another archivethat contains layers of reference vectors corresponding to different density.Experimental results show the expected characteristics and competitiveperformance of the proposed algorithm TEEA.