2200 2400 2600 2800 3000 3200 3400 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 values objectives Are All Objectives Necessary? On Dimensionality Reduction in Evolutionary Multiobjective Optimization Dimo Brockhoff and Eckart Zitzler Computer Engineering Computer Engineering and Networks Laboratory and Networks Laboratory Results 0 10 20 30 40 50 5 10 15 20 2 4 6 8 10 12 14 16 number of objectives needed exact greedy delta k number of objectives needed 0 10 20 30 40 50 5 10 15 20 10 1 10 2 10 3 10 4 10 5 10 6 10 7 10 8 10 9 runtimes [ms] exact greedy delta k runtimes [ms] entire search space of 0-1-knapsack problem with 7 items Exact algorithm vs. heuristic Different problems act differently Pareto front approximations for 0-1-knapsack and DTLZ7 Motivation Multiobjective Problem EA Approximation of Pareto front which solution is best? reduce number of objectives Assist decision maker error: δ = 0 Conclusions Approach heuristic slightly worse results, but clearly faster ⇒ Benefits of the Approach: • Definition of conflict between objective sets can detect redundancy • Objective reduction is adjustable by defining error threshold or largest allowed objective set size • Approach guarantees maximal error in dominance structure change Take Home Message: Given a set of solutions, objective reduction is possible by preserving or only slightly changing the dominance structure. The omission of redundant information can assist the decision maker. Key Contributions: • Generalization of conflict between objective sets • Framework for objective reduction to assist the decision maker ⇒ the smaller the objective set, the larger the error general statements on redundancy impossible ⇒ δ−MOSS k-EMOSS 0 5 10 15 20 25 30 35 40 20 10 0 number of objectives needed delta [% of population’s spread] knapsack, 15 objectives knapsack, 25 objectives DTLZ7, 15 objectives DTLZ7, 25 objectives 0 0.2 0.4 0.6 0.8 1 90 60 30 error in computed set k [% of entire objectives] knapsack, 15 objectives knapsack, 25 objectives DTLZ7, 15 objectives DTLZ7, 25 objectives Problem: Decision making with many objectives is challenging Questions: • Can objectives be omitted while the dominance structure is preserved/only slightly changed? • How to compute a minimum objective set? 2500 2600 2700 2800 2900 3000 3100 3200 19 14 4 2 1 values objectives Drawbacks of known dimensionality reduction approaches: • Not suitable for black-box optimization [Agrell 1997] • No guarantee to preserve dominance structure [Deb and Saxena 2005] Algorithms exact • , and resp., for δ-MOSS and k-EMOSS greedy • for δ-MOSS • for k-EMOSS The Miminum Objective Subset Problems Given: Solution set with objective values δ−MOSS: Compute a minimum objective set, yielding a slightly changed relation with error δ -EMOSS: Compute an objective set with objectives, changing the relation least Objective Conflicts Preservation of dominance structure • Pairwise objective conflicts non-redundancy • Omission of objectives possibly additional edges dominance structure preserved slightly changed Dimensionality Reduction Changes in dominance structure omit omit and 1 3 2 4 obj. values obj. values