Data-efficient optimization of multifunctional polymer systems remains challenging because of high-dimensional formulation spaces and competing property requirements. Here, we present a multi-objective Bayesian optimization (BO) framework for accelerated design of electroactive polymer (EAP) formulations, demonstrated on PEDOT:PSS/PVA composite films. A mixed-level combinatorial seed (50 formulations) spanning base composition, solvent doping, and multiple additive families was experimentally realized and characterized for electrical and mechanical performance. Experimental data were curated through a quality-controlled, uncertainty-aware pipeline and encoded as constrained descriptors for mixed-variable Gaussian process regression surrogate modeling.
Benchmarking on the Branin-Currin function demonstrates that batch multi-objective BO using q-expected hypervolume improvement (qEHVI) substantially outperforms random sampling in convergence rate and Pareto front identification. Applied to experimental data, the model captures physically consistent structure-property relationships, and identifies solvent additives and plasticizers as key drivers of mechanical compliance while preserving conductivity. The evolving Pareto front reveals pronounced trade-offs between stiffness, ductility, and electrical performance, underscoring the necessity of multi-objective optimization. This work establishes a reproducible, closed-loop framework for data-efficient exploration of complex polymer formulation spaces.