They get it! New ‘machine unlearning’ technique deletes unwanted data

Bam!

The novel approach to making systems forget data is called “machine unlearning” by the two researchers who are pioneering the concept. Instead of making a model directly depend on each training data sample (left), they convert the learning algorithm into a summation form (right) – a process that is much easier and faster than retraining the system from scratch.

  1. See! “Summation” first, then details!!!!! What they are really doing is putting things in a fractal structure.
  2. And they are understanding that many aspects of “old” data have little value.
Advertisements

Leave a Reply

Please log in using one of these methods to post your comment:

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