在情感认知的学习与决策中引入了情绪认知评价理论,提出了基于情绪认知评价理论的人机交互情感决策,对情感行为的选取进行了优化;在情感迷宫模型中,对该决策算法进行了Matlab仿真试验,试验结果表明使用BpQ-learning算法的智能体在寻找目标情感过程中得到的平均奖励值高、试探的次数少,达到了预期的试验目标。
Emotions play an important role in the daily lives of people study and work, such as reasoning, learning, memory, and decision-making. In computers and artificial intelligence area, reserches have found that it is not enough for intelligent machines to only have the power of logic computing, it should also be given emotional competence. In order to achieve computer intelligent emotional interaction, the computer must have human-like emotions that can interact with peo ple. In this paper, a new emotional decision-making algorithm was presented for affective agents inspired by cognitive evalution theory of emotion, and the emotional behavior selection was opti mized. This algorithm was applied to the emotional maze model, Matlab experimental results show that the use of BpQ-learning algorithm to find the target in a virtual affective agents had high average reward value and less number of trials.