Tag
This paper presents a multi-objective Bayesian optimization approach to automate weight selection in reinforcement learning for energy-aware control, demonstrating superior sample efficiency over grid search on a physical Quanser Aero 2 testbed.
Introduces Partition-Guided Distance Saliency (PGDS), a novel XAI framework for many-objective optimization that uses geometric intuition to explain how decision variables influence objective space proximity, validated on 10-objective benchmarks and a physics-informed engineering problem.
CRAFT is a Pareto-front prompt optimizer that jointly optimizes for accuracy and token cost, avoiding the 'scalarization collapse' of weighted-sum approaches by maintaining a diverse population of prompts across the accuracy-cost trade-off frontier using NSGA-II and budget-aware validation.