Got it

Feature Learning - an introduction

Latest reply: Mar 15, 2021 13:13:14 138 8 10 0 0

Good day Huawei family!


After our previous blog on Deep Learning, I thought it would be a good idea to delve deeper into data-related topics and feature one about... Feature Learning.


An extremely common characteristic of Machine Learning, Feature Learning has been fascinating both practitioners and data scientists for quite some time now. Let us find out more about Feature Learning in the upcoming sections of this blog!


BACKGROUND INFORMATION


Please watch the below video, which I believe conveys very well what the concept of Feature Learning really is about:



This begs for a written definition, does it not? That's what I thought. Say no more, as thanks to Stanford University's online Unsupervised Feature Learning and Deep Learning Tutorial we can get a pretty good one: 'if we can get our algorithms to learn from ”unlabeled” data, then we can easily obtain and learn from massive amounts of it'.


The Tutorial goes even further with the explanation, highlighting the exquisite properties of Feature Learning: 'even though a single unlabeled example is less informative than a single labeled example, if we can get tons of the former - for example, by downloading random unlabeled images/audio clips/text documents off the internet - and if our algorithms can exploit this unlabeled data effectively, then we might be able to achieve better performance than the massive hand-engineering and massive hand-labeling approaches'.


So, technically, to wrap everything up, Feature Learning is nothing but a bundle of techniques which learn a feature. In other words, this is a transformation of raw data input which can be exploited efficiently in the tasks of Machine Learning. Pretty cool concept, don't you think?


WHERE CAN FEATURE LEARNING BE USED?


images-3853117_1920

There are two types of Feature Learning: supervised feature learning and unsupervised feature learning. Therefore, please find below the scenarios that can be found in each category.

Supervised feature learning scenarios:

  • supervised dictionary learning;

  • neural networks.

Unsupervised feature learning scenarios:

  • K-means clustering;

  • principal component analysis;

  • local linear embedding;

  • independent component analysis;

  • unsupervised dictionary learning.

FEATURE LEARNING BENEFITS


stockvault-human-biometrics258196


Feature Learning provides countless benefits to the enterprise that decides to go with this approach. It is worth, thus, mentioning:


  • reducing the large input of data for an algorithm to be processed - building derived values like TF-IDF (features) to be informative (rich of information) and non-redundant;

  • learning transformations of data that make it easier to extract useful information - especially when building classifiers or other predictors.

THE BOTTOM LINE


What more could it be said but machines are indeed humanity's pinnacle invention and the fact they can teach themselves things without human supervision only translates as high computational power? If you agree with this statement, please go ahead and subscribe to our Community blog - there will always be more interesting content on new technologies coming up every week!

The post is synchronized to: Community Blog

  • x
  • convention:

little_fish
Admin Created Feb 26, 2021 13:37:24

good
View more
  • x
  • convention:

andersoncf1
Created Feb 26, 2021 19:25:11

Nice
View more
  • x
  • convention:

Network%20Engineer%2C%20Post-Graduation%20in%20Computer%20Networks%20and%20Security%2C%20Possessing%20Certifications%20of%20Huawei%3A%20HCIE-R%26S%20(Written)%2C%202x%20HCIP-R%26S%20and%20Security%20and%203x%20HCIA-R%26S%2C%20Security%20and%20WLAN.
Kevin_Thomas
Created Feb 27, 2021 10:31:03

Nice!
View more
  • x
  • convention:

Kevin_Thomas
Kevin_Thomas Created Feb 27, 2021 10:31:11 (0) (0)
 
nagu
Created Mar 10, 2021 08:34:36

Thanks for sharing
View more
  • x
  • convention:

Unicef
MVE Created Mar 10, 2021 13:03:29

Good sharing
View more
  • x
  • convention:

Live%20Lead%20Love%20%3A)
Faridrami
Created Mar 10, 2021 16:11:38

Thanks for sharing
View more
  • x
  • convention:

NTan33
Created Mar 15, 2021 13:13:14

Who knows what interesting things future developments in machine learning can result in?
View more
  • x
  • convention:

Someone%20new%20and%20curious%20about%20IT.

Comment

You need to log in to comment to the post Login | Register
Comment

Notice: To protect the legitimate rights and interests of you, the community, and third parties, do not release content that may bring legal risks to all parties, including but are not limited to the following:
  • Politically sensitive content
  • Content concerning pornography, gambling, and drug abuse
  • Content that may disclose or infringe upon others ' commercial secrets, intellectual properties, including trade marks, copyrights, and patents, and personal privacy
Do not share your account and password with others. All operations performed using your account will be regarded as your own actions and all consequences arising therefrom will be borne by you. For details, see " Privacy."

My Followers

Login and enjoy all the member benefits

Login

Block
Are you sure to block this user?
Users on your blacklist cannot comment on your post,cannot mention you, cannot send you private messages.
Reminder
Please bind your phone number to obtain invitation bonus.