An Empirical Systematic Study of Deep Learning-Based Software Defect Prediction Models in Complex Systems
Yang Chen, Min Fang, Yang Zhang · 2024
In this paper, the advanced reinforcement learning technology is integrated, and an innovative strategy of teaching resources optimization and students' learning path adaptive adjustment is proposed. By using the instant feedback and strategy update ability of the Actor-Critic algorithm, and the decision-making advantage of DQN in high-dimensional state space, the accurate recommendation of students personalized learning paths and the efficient allocation of teaching resources are realized. In this paper, the strategy network (Actor) is constructed to explore and formulate the learning path strategy, the value network (Critic) is used to evaluate the quality of the strategy, and the rapid strategy iteration is realized. The Q-value function is constructed based on the deep neural network, which is used to select the optimal or near-optimal learning resource allocation scheme among a large number of possible actions, so as to improve students' learning effectiveness and satisfaction. Optimizing the management efficiency of practical courses provided a new perspective and feasible scheme for the future development of smart education.