Abstract:

This paper is focused on the impact of the innovative framework Multiple Intelligences Agent Deep Q-Network (MIADQN). MIADQN incorporates the Theory of Multiple intelligences, developed by Howard Gardner and deep reinforcement learning. Gardner’s theory identifies eight types of intelligence- linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, interpersonal, intrapersonal and naturalistic. Each of these intelligences represents a unique cognitive processing style and therefore MIADQN functions as a multi-agent system, where each agent embodies one of these intelligence types. The multi-agents handle real-time data from a dynamically changing environment filled with interactive activities to map individual learning preferences and to uncover cognitive strengths. Through deep reinforcement learning, MIADQN adjusts and enhances content strategies and refines tasks to optimize learning outcomes, ultimately creating personalized users` profiles. It leverages Deep Q-networks to guide agents using cumulative rewards tied to user engagement and performance metrics. These agents constantly improve their performance by balancing exploration with exploitation during the reinforcement learning process. This adjustability enables the system to modify tasks and materials in real time, aligning them with the unique profile of each user it interacts with. MIADQN addresses the challenges of personalized education by suggesting tailored learning pathways and complementing educators with valuable insights. The paper outlines MIADQN’s architecture, and demonstrates its potential to merge artificial intelligence with real-world classroom applications. Ultimately, MIADQN provides a scalable and practical solution for enhancing personalization in education through the usage of artificial intelligence.

Keywords:

Multiple Intelligences, Deep Reinforcement learning, Education, Technology, Personalized learning.