saealib.WeightedSumDecomposition#

class saealib.WeightedSumDecomposition[source]#

Bases: Decomposition

Linear weighted-sum scalarization.

score_i = w · f_i

The ideal point is not used. This is the simplest decomposition but cannot reach non-convex parts of the Pareto front — use TchebycheffDecomposition or PBIDecomposition for non-convex problems.

References

[9]: Zhang, Q., & Li, H. (2007). MOEA/D: A multiobjective evolutionary algorithm based on decomposition. IEEE Transactions on Evolutionary Computation, 11(6), 712-731.

Methods

`__init__`
`aggregate`	Compute the weighted dot product `f @ weights`.
`contract`	Return the decomposition contract.

Method Details

WeightedSumDecomposition.__init__()#

WeightedSumDecomposition.aggregate(f, weights, ideal_point)[source]#

Compute the weighted dot product f @ weights.

Parameters:

f (np.ndarray) – Objective matrix, shape (N, n_obj).
weights (np.ndarray) – Non-negative weight vector, shape (n_obj,).
ideal_point (np.ndarray) – Ignored.

Returns:

Scores, shape (N,).

Return type:

np.ndarray

WeightedSumDecomposition.contract()#

Return the decomposition contract.

Return type:: ComponentContract