Correctness Learning: Deductive Verification Guided Learning for Human-AI Collaboration
This work addresses verification challenges in human-AI collaboration for safety-critical applications, representing an incremental advancement by combining existing methods in a novel way.
The paper tackles the problem of verifying correctness in AI decision-making for safety-critical fields by proposing correctness learning (CL), which integrates deductive verification with historical high-quality schemes to guide intelligent agents, resulting in improved decision-making and resource optimization as validated through extensive experiments.
Despite significant progress in AI and decision-making technologies in safety-critical fields, challenges remain in verifying the correctness of decision output schemes and verification-result driven design. We propose correctness learning (CL) to enhance human-AI collaboration integrating deductive verification methods and insights from historical high-quality schemes. The typical pattern hidden in historical high-quality schemes, such as change of task priorities in shared resources, provides critical guidance for intelligent agents in learning and decision-making. By utilizing deductive verification methods, we proposed patten-driven correctness learning (PDCL), formally modeling and reasoning the adaptive behaviors-or 'correctness pattern'-of system agents based on historical high-quality schemes, capturing the logical relationships embedded within these schemes. Using this logical information as guidance, we establish a correctness judgment and feedback mechanism to steer the intelligent decision model toward the 'correctness pattern' reflected in historical high-quality schemes. Extensive experiments across multiple working conditions and core parameters validate the framework's components and demonstrate its effectiveness in improving decision-making and resource optimization.