An Approach to Interactive Affective Learning Algorithms

To solve multi-objective decision-making problems without explicit mathematical description for objective functions, traditional interactive evolutionary computing approaches are usually limited in searching ability and vulnerable to human’s subjectivity. Motivated by this observation, a novel affective computing and learning solution adapted to human-computer interaction mechanism is explicitly proposed. Therein, a kind of stimulating response based affective computing models (STAM) is constructed, along with quantitative relations between affective space and human's subjective preferences. Thereafter, affective learning strategies based on genetic algorithms are introduced which are responsible for gradually grasping essentials in human’s subjective judgments in decision-making, reducing human’s subjective fatigue as well as making the decisions more objective and scientific. To exemplify applications of the proposed methods, test functions are suggested to case studies, giving rise to satisfied results and showing validity of the contribution.