This study explores the memory characteristics of elderly individuals to design effective smart devices based on smart memory storage solutions under deep learning to improve the learning efficiency of elderly individuals. The different memory formation stages in the existing human brain are analysed. A smart memory storage solution based on memory-enhanced embedded learning is constructed based on meta-learning under deep learning, which reduces the cost of learning new tasks to the greatest extent. Finally, the performance of the proposed solution is verified using different datasets. The results reveal that the solution based on deep learning has obvious effects on different datasets, with an average accuracy rate of 99.7%. By synthesizing a large number of target sample features, this solution can lower the learning difficulty and improve the learning effect. The proposed elderly oriented smart device effectively reduces the shortcomings in the current market and lowers the learning difficulty, which provides an important reference for further enriching devices in the ageing market.
展开▼
