TY - GEN
T1 - Tree-based genetic programming for evolutionary analog circuit with approximate Shapley value
AU - Shi, Xinming
AU - Minku, Leandro L.
AU - Yao, Xin
PY - 2024/11/29
Y1 - 2024/11/29
N2 - The automated design of analog circuits presents a significant challenge due to the complexity of circuit topology and parameter selection. Traditional evolutionary algorithms, such as Genetic Programming (GP), have shown potential in this domain but are often hindered by inefficient search processes and the large design space. Furthermore, fitness evaluation in the evolutionary design of circuits is often computationally very expensive. In this paper, we introduce a novel evolutionary framework that leverages approximate Shapley values to guide the optimization process in tree-based genetic programming for analog circuit design. Our approach addresses the computational challenges associated with computing Shapley values by introducing a two-stage evolutionary framework that includes a Shapley Value Library (SVlib) and a KNNbased prediction for efficient estimation of Shapley values. Our proposed work not only enhances the search efficiency by focusing on the most beneficial subcircuits but also leads to more compact and efficient circuit designs. Furthermore, fitness evaluation in the evolutionary design of circuits is often computationally very expensive experiments, we verify that our framework accelerates evolutionary convergence and outperforms traditional methods in terms of circuit optimization.
AB - The automated design of analog circuits presents a significant challenge due to the complexity of circuit topology and parameter selection. Traditional evolutionary algorithms, such as Genetic Programming (GP), have shown potential in this domain but are often hindered by inefficient search processes and the large design space. Furthermore, fitness evaluation in the evolutionary design of circuits is often computationally very expensive. In this paper, we introduce a novel evolutionary framework that leverages approximate Shapley values to guide the optimization process in tree-based genetic programming for analog circuit design. Our approach addresses the computational challenges associated with computing Shapley values by introducing a two-stage evolutionary framework that includes a Shapley Value Library (SVlib) and a KNNbased prediction for efficient estimation of Shapley values. Our proposed work not only enhances the search efficiency by focusing on the most beneficial subcircuits but also leads to more compact and efficient circuit designs. Furthermore, fitness evaluation in the evolutionary design of circuits is often computationally very expensive experiments, we verify that our framework accelerates evolutionary convergence and outperforms traditional methods in terms of circuit optimization.
U2 - 10.1007/978-3-031-77915-2_18
DO - 10.1007/978-3-031-77915-2_18
M3 - Conference contribution
T3 - Lecture Notes in Computer Science
SP - 253
EP - 267
BT - Forty-fourth SGAI International Conference on Artificial Intelligence: Proceedings
PB - Springer
T2 - 44th SGAI International Conference on Artificial Intelligence 2024
Y2 - 17 December 2024 through 19 December 2024
ER -