Deep Machine Learning - A New Frontier in Artificial Intelligence Research [Research Frontier]
Глубокое машинное обучение — новый рубеж исследований в области искусственного интеллекта [Перспективное направление исследований]
2010-10-20
SCID: 54.1/r68h3dbu
Discuss with AI
artificial intelligenceconvolutional neural networksdeep belief networksdeep learningdeep machine learning
Figures from the paper
Abstract (AI)
This article provides an overview of the mainstream deep learning approaches and research directions proposed over the past decade. It is important to emphasize that each approach has strengths and "weaknesses, depending on the application and context in "which it is being used. Thus, this article presents a summary on the current state of the deep machine learning field and some perspective into how it may evolve. Convolutional Neural Networks (CNNs) and Deep Belief Networks (DBNs) (and their respective variations) are focused on primarily because they are well established in the deep learning field and show great promise for future work.
Key Findings
1
Convolutional Neural Networks and Deep Belief Networks receive primary attention because they are established approaches with strong future research potential.
2
Deep learning methods have application- and context-dependent strengths and weaknesses, requiring appropriate selection for specific uses.
3
The article reviews mainstream deep learning approaches and major research directions developed over the past decade.
4
The article summarizes the current state of deep machine learning and offers perspectives on its likely evolution.
Research Object
deep machine learning approaches, particularly Convolutional Neural Networks (CNNs) and Deep Belief Networks (DBNs)
Research Subject
their strengths, weaknesses, current state, and future research directions across applications and contexts
Publication Details
Publication Date
2010-10-20
Journal
Publisher
ISSN
Cited by
1255
Access Type
Author Information
Download PDF
Subscribe to digest