Machine learning in practice

In the last week of my old job, I saw a talk from one of Facebook’s Engineers about how they use machine learning in practice. His talk boiled down to 4 points:

  • More data, better quality data: spend time collecting and cleaning data
  • Practice != theory: often simple models work better in practice as better ones may be too slow
  • Efficiency is key: getting something to work in real time with lots of data is hard
  • 99% Perspiration: actually running the classifier is a tiny fraction of the time

 

About these ads
This entry was posted in Machine Learning and tagged , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s