Genetic algorithm and binary masks for co-learning multiple datasets in deep neural networks

This study addresses the challenges of 'catastrophic forgetting' and 'multi-task learning' encountered in the field of data classification and analysis, particularly with the use of Convolutional Neural Networks (CNNs). The aim of the study is to use genetic algorithms to solve these problems. Methodologically, an optimization strategy has been developed that employs layer-based binary masks to customize CNNs models for multiple datasets. Genetic algorithms have been utilized as a heuristic search method to optimize a binary mask for each dataset. Experiments have been conducted on widely-used dataset such as MNIST, Fashion MNIST, and KMNIST. The obtained results are notably impressive, with classification accuracies of 76.25% for MNIST, 76% for Fashion MNIST, and 74.43% for KMNIST. These findings demonstrate that the proposed approach can create high-performance models not only for a single task but also for multiple tasks.