Special issue on deep learning-based neural information processing for big data analytics
Chuanchao Huang
DOI: https://doi.org/10.1007/s00521-019-04602-1
2019-11-04
Neural Computing and Applications
Abstract:Living in the era of big data, we have been witnessing the dramatic growth of heterogeneous data, which consists of a complex set of cross-media content, such as text, images, videos, audio, graphics, and time series sequences. Such hybrid data comes from multiple sources and hence embodies different feature spaces. This situation is creating new challenges for the design of effective algorithms and developing generalized frameworks to meet heterogeneous computing requirements. Meanwhile, deep learning is revolutionizing diverse key application areas, such as speech recognition, object detection, image classification, and machine translation, with its data-driven representation learning. Thus, it has become critical to explore advanced deep learning techniques for heterogeneous big data analytics, including data acquisition, feature representation, time series analysis, knowledge understanding, and semantic modeling. With the resurgence of deep learning architectures and learning methods, neural information processing has been applied to a variety of disciplines and proved highly successful in a vast class of applications. For instance, in the pattern recognition field, deep neural networks achieved human-like performance in recognizing, labeling and sorting images, e.g., on the ImageNet benchmark. Moreover, DeepMind's AlphaGo Zero, trained by self-play reinforcement learning, achieved superhuman performance in the game of Go. On the hardware architecture side, advanced neuromorphic processors have been designed to mimic human functions of perception, motor control and multisensory integration.
computer science, artificial intelligence