Source-linked AI summary

Review of Deep Learning

Rong Zhang, Weiping Li, Tong Mo

arXiv:1804.01653v2cs.LGcs.CVcs.NEstat.ML

TL;DR

Deep learning is a key area of artificial-intelligence research amid increasing investment. This paper reviews its basic and emerging models, applications, research progress, future directions, and existing problems.

  • Problem

    Deep learning has become a key area of artificial-intelligence research as countries and high-tech companies increase investment in artificial intelligence.

  • Method

    The paper outlines multilayer perceptrons, convolutional neural networks, and recurrent neural networks, then examines convolutional and recurrent-network variants.

  • Results

    The paper summarizes recent deep-learning progress, future research directions, and applications across artificial-intelligence areas.

  • Takeaways & Limitations

    The review organizes deep learning around foundational models, emerging model developments, applications, and possible solutions to existing problems.

  • Takeaways & Limitations

    Recurrent neural networks have gradient vanishing or gradient explosion problems, while deep learning is also vulnerable to adversarial-sample attacks.

Abstract

from arXiv · show

In recent years, China, the United States and other countries, Google and other high-tech companies have increased investment in artificial intelligence. Deep learning is one of the current artificial intelligence research's key areas. This paper analyzes and summarizes the latest progress and future research directions of deep learning. Firstly, three basic models of deep learning are outlined, including multilayer perceptrons, convolutional neural networks, and recurrent neural networks. On this basis, we further analyze the emerging new models of convolution neural networks and recurrent neural networks. This paper then summarizes deep learning's applications in many areas of artificial intelligence, including speech processing, computer vision, natural language processing and so on. Finally, this paper discusses the existing problems of deep learning and gives the corresponding possible solutions.

Loading 1804.01653v2…