"Reprint" Deep Learning & Neural Network Popular Science and gossip study notes

The previous article mentions the difference between data mining, machine learning, and deep learning: http://www.cnblogs.com/charlesblc/p/6159355.html

Deep learning specific content can be seen here:

Refer to this article: Https://zhuanlan.zhihu.com/p/20582907?refer=wangchuan "Wang Chuan: How deep is the depth of learning, how much did you learn?" (i) "Note: Neural network research, because the artificial intelligence of a Daniel Marvin Minsky not optimistic, and the book shows its limitations, and 20 years of long-term low tide." The second: Https://zhuanlan.zhihu.com/p/20592962?refer=wangchuan Notes: Distributed Characterization (distributed representation) is a core idea of neural network research.

"White Truck", if distributed, a neuron represents the size, one neuron represents the color, and the third neuron represents the category of the vehicle. When three neurons are active at the same time, we can accurately describe the object we want to express.

Distributed characterization, compared to traditional local characterization (localized representation), has a much higher storage efficiency. The number of neurons that are linearly increased can be expressed in a number of different concepts that increase exponentially.

Another advantage of distributed characterization is that the expression of information is not fundamentally compromised, even in the event of a local hardware failure.

This idea let Geoffrey Hinton Epiphany, so that he has been in the field of neural network research for more than 40 years has not flinched.

A generation of neural networks is the Geoffrey Hinton of the University of Toronto. Geoffrey Hinton and Wunda with Google for Deep learning (Google Brain Project). Let's take a picture, from https://www.zhihu.com/question/27177122:

From left to right:

Yann LeCun: convolutional neural Network, reference http://www.leiphone.com/news/201406/deep-learning-yann-lecun-facebook.html
Geoffrey Hinton: The originator of deep learning
Yoshua Bengio: Also deep learning
Andrew Ng:hinton's partner

All are deep learning related, can also read this article: Link "Yann LeCun, Geoffrey Hinton or Yoshua Bengio can get Turing award?" 》

These people also have intricate relationships:

In the series, can find all articles: Https://zhuanlan.zhihu.com/wangchuan His public number of articles is also very good, you can see: Link

"Reprint" Deep Learning & Neural Network Popular Science and gossip study notes