Surrogate-assisted multi-objective evolutionary algorithms have advanced the field of computationally expensive optimization, but their progress is often restricted to lowdimensional problems. This manuscript presents a multiple classifiers-assisted evolutionary algorithm based on decomposition, which is adapted for high-dimensional expensive problems in terms of the following two insights. Compared to approximationbased surrogates, the accuracy of classification-based surrogates is robust for few high-dimensional training samples. Further, multiple local classifiers can hedge the risk of over-fitting issues. Accordingly, the proposed algorithm builds multiple classifiers with support vector machines on a decomposition-based multiobjective algorithm, wherein each local classifier is trained for a corresponding scalarization function. Experimental results statistically confirm that the proposed algorithm is competitive to the state-of-the-art algorithms and computationally efficient as well.

No abstract

Slow light generated through silicon (Si) photonic crystal waveguides (PCWs) is useful for improving the performance of Si photonic devices. However, the accumulation of coupling loss between a PCW and Si optical wiring waveguides is a problem when slow-light devices are connected in a series in a photonic integrated circuit. Previously, we reported a tapered transition structure between these waveguides and observed a coupling loss of 0.46 dB per transition. This Letter employed particle swarm optimization to engineer the arrangement of photonic crystal holes to reduce loss and succeeded in demonstrating theoretical loss value of 0.12 dB on average in the wavelength range of 1540–1560 nm and an experimental one of 0.21 dB. Crucially, this structure enhances the versatility of slow light.

No abstract

No abstract