Search

In today’s booming development of free trade ports, China, as a big country of inbound and outbound tourism, has put forward higher requirements for the talent cultivation mode of foreign-related tourism English. This paper takes the characteristics of foreign-related tourism English as the entry point and establishes the practical teaching system and the four-in-one talent cultivation mode of foreign-related tourism English. The multi-objective optimization model for cultivating foreign tourism English talents is proposed, with the objective function of enhancing the quality of foreign tourism English talents, and the corresponding constraints are set. The multi-objective optimization model is solved using the DQN model, and a strategy gradient algorithm-based model training optimization strategy is proposed. The test conditions of foreign-related tourism English talent cultivation are set, simulation experiments are carried out, and teaching experiments are compared based on the optimal scheme. The training loss obtained by the DQN model after 1700 iterations fluctuates between [0.015,0.019], and when the training period is set to 2 years, the cumulative return of foreign-related tourism English talent cultivation can reach 623.87%, and the maximum retracement is only - 5.457%. The optimal talent cultivation program’s average score is 14.1 points higher than the ordinary version. Foreign-related tourism English needs to be fully integrated with the actual demand for talent, and the faculty and policy guidelines help to enhance the quality of foreign-related tourism English talent training.

This paper explores the principle of deep reinforcement learning algorithm application and constructs the digital teaching model of Civics through the recurrent neural network, Markov decision process, Actor-Critic algorithm and collaborative filtering recommendation algorithm. On this basis, the state representation model and decision-making model are invoked to improve the diversity optimization recommendation of the deep reinforcement learning algorithm. Aiming at the problems existing in the practical application of the OBE concept, the model constructed in this paper is utilized to propose a multi-platform integration of the Civics digital learning model. The objective of empirical and simulation experiments is to confirm the model’s implementation and recommendation effects, respectively. In the validation of hyperparameter d and N values, the DQN model achieves the optimal recommendation effect when the d values are 65 and 63 on the HetRec and MovieTweetings datasets. During the implementation of the digital teaching evaluation of the course Civics, more than 60% of the students rated the teacher’s feedback session higher. Overall, the 6 dimensions were rated higher, and the number of students who chose to be fully compliant and conforming was more than 200.

In this study, a new blurred line enhancement scheme of computer art image based on Deep Reinforcement Learning (DRL) algorithm was proposed. The hard threshold method is adopted to remove the noise of computer art images and the blurred lines are extracted by texture separation method. Based on the line extraction results, the Deep Q-Network (DQN) model was built with DRL algorithm, and the sample images were input into the model, and the fuzzy line enhancement results of computer art images were obtained in the output layer. The proposed method exhibits excellent noise reduction effect, and the fuzzy line enhancement quality of computer art image is good. The average enhancement time is 0.58 s, and the practical application effect is good. [ABSTRACT FROM AUTHOR]

• In the empirical analysis, the research model has excellent anti-interference ability, and the error is relatively smaller. • The overall satisfaction score of the students on the system reaches more than 90 points, effectively improving the learning effect and academic performance. Training game intelligence through reinforcement learning (RL) and applying games to teaching and learning can stimulate students' enthusiasm for learning. Therefore, the study built an online game teaching intelligence platform based on deep RL to enhance students' interest and enthusiasm in independent learning and improve teaching effectiveness through the game model. The study showed that the research model, after incorporating the Softplus function, had a higher overall score compared to other models in Breakout game test after 50 game simulations, with the highest average score of 3.2. In the Flappy Bird game, the research model had the fastest growth rate and the highest increase in score, with a final score of over 120 and an average score of 63.1 for the game after the improvement of the activation function. When compared with the Double-Deep Q-network (D-DQN) model, the research model scored higher performance after training and the model curve performed better in convergence. The research model suffered less interference in noise amplitude and frequency, and performed better in adjusting the weights of the neural network. In the empirical analysis, the research model had excellent interference resistance and relatively smaller errors. The overall satisfaction score of students with the system reached over 90 points, effectively improving learning effectiveness as well as academic performance. [ABSTRACT FROM AUTHOR]