Implementation of different encoding types on structural optimization based on adaptive genetic algorithm

Multi-objective mapping optimization via problem decomposition for many-core systems

“…Thus, designers usually have to deal with paretooptimal sets rather than single solutions. A simple approach is to aggregate all objectives into a single, parameterized objective function [3]. Though this approach is beneficial in the sense that well-established single-objective technique can be used seamlessly, it is sensitive to the shape of the pareto front, prior knowledge on the problem may be required, and it requires several optimization runs which do not exploit synergy from each other.…”

“…To properly manage populations to evolve to a pareto-optimal set, nondominated sorting [5], strength pareto evolutionary algorithm [26], and simple indicator algorithm [25] have been proposed. Problem specific specialization of operators (i.e., crossover and mutation) and different chromosomal representations have also been studied [3], [7], [9]. However, MOEAs are still not scaling well to large problems.…”

“…For instance, L p norm of processor utilization in state s is uc s (s) = L p ({uc cs (c, s)|∀c ∈ C}) and the weighted sum of all states is uc = s∈S w s · uc s (s). ul (2) and ul (3) can be calculated in a similar way. In summary, the objective vector f to be minimized is f = (uc, ul (2) , ul (3) ).…”

“…ul (2) and ul (3) can be calculated in a similar way. In summary, the objective vector f to be minimized is f = (uc, ul (2) , ul (3) ).…”

“…To take a deeper look at the effect of the architecturebased decomposition, we project the obtained pareto curves of SynthBig from EXPO and SADEXPO onto the 2-dimensional space of uc and ul (3) as shown in Fig. 9.…”