binary_indicator function

Binary quality indicators

Calculates the quality indicator value of the set of points given in x with respect to the set given in o. As with all functions in emoa that deal with sets of objective values these are stored by column.

hypervolume_indicator(points, o, ref)

epsilon_indicator(points, o)

r1_indicator(points, o, ideal, nadir, lambda, utility = "Tchebycheff")

r2_indicator(points, o, ideal, nadir, lambda, utility = "Tchebycheff")

r3_indicator(points, o, ideal, nadir, lambda, utility = "Tchebycheff")

Arguments

• points: Matrix of points for which to calculate the indicator value stored one per column.
• o: Matrix of points of the reference set.
• ref: Reference point, if omitted, the nadir of the point sets is used.
• ideal: Ideal point of true Pareto front. If omited the ideal of both point sets is used.
• nadir: Nadir of the true Pareto front. If ommited the nadir of both point sets is used.
• lambda: Number of weight vectors to use in estimating the utility.
• utility: Name of utility function.

Returns

Value of the quality indicator.

References

Zitzler, E., Thiele, L., Laumanns, M., Fonseca, C., and Grunert da Fonseca, V (2003): Performance Assessment of Multiobjective Optimizers: An Analysis and Review. IEEE Transactions on Evolutionary Computation, 7(2), 117-132.

Author(s)

Olaf Mersmann olafm@statistik.tu-dortmund.de