This interesting paper summarizes work for the incremental improvement of automatic analog design and optimization, by leveraging capabilities of multiple existing leading algorithms. Although the proposed algorithm is an aggregate of a few existing ones, Liu et al. manage to develop a more powerful method than the old ones. The comparison between the new memetic single-objective evolutionary algorithm (MSOEA) and the legacy ones--such as genetic algorithms and penalty functions (GAPF), differential evolution and penalty functions (DEPF), the co-evolutionary differential evolution algorithm (CODE), and the selection-based differential evolution algorithm (SBDE)--is convincing, showing promising results for the new algorithm.

One weakness of the paper is that the chosen examples are mostly primitive analog building blocks with small design size, so whether the new algorithm will hold its edge when scaled to larger designs is not conclusive. Scalability is one of the most important factors in automatic circuit designs. Therefore, the authors should continue this work on more realistic, larger designs.