ABSTRACT
Today, deep learning is perhaps the most widely used machine learning topic. Deep learning models have indeed been prominent in various domains in recent times, including computer vision, satellite image processing, recommendation systems, health-care systems, and processing of natural language. While extensive data provides a sufficient number of training items, it also poses a difficulty to deep learning. As a result, several models for considerable learning of data have been created in recent years and we describe deep learning methods for learning of big data from four perspectives: multi-modal deep learning models, deep computation models, incremental deep learning models, and reliable deep learning models.
