Abstract: Due to the significant differences in the chemical and physical properties of aluminum and copper, numerous defects can arise at the welded joint during the welding process, severely affecting the joint's performance. High-entropy alloys provide significant advantages in enhancing the welding performance of dissimilar metals. Therefore, this study employed FeCoNiCrTi high-entropy alloy powder as a filler material. Additionally, the study proposed an optimization method for welding parameters using a gradient boosting regression tree model optimized by the black-winged kite algorithm, combined with a multi-objective stochastic paint optimizer. The results indicate that under the conditions of a laser power of 677.2 W, a welding speed of 639.3 mm/min, a defocusing amount of 2.75 mm, and a high-entropy alloy addition of 0.05 g, the optimization objectives reached optimal levels. The warp deformation of the welded part decreased by 20.14%, the maximum tensile strength of the welded part increased by 49.72%, and the cost was reduced by 10.90%.