Barber, David;
(2023)
A note on Double Descent.
(Research Note
).
UCL Centre for Artificial Intelligence: London, UK.
Preview |
Text
DoubleDescent (1).pdf - Submitted Version Download (213kB) | Preview |
Abstract
Double Descent is the phenomenon that the test error in a learning system displays non-monotonic behaviour as the number of train datapoints increases. Double Descent was well known in the 1990s and this brief note adds some references and details around how to calculate the test error and suggests an explanation for the phenomenon.
| Type: | Working / discussion paper |
|---|---|
| Title: | A note on Double Descent |
| Open access status: | An open access version is available from UCL Discovery |
| Publisher version: | http://web4.cs.ucl.ac.uk/staff/D.Barber/DoubleDesc... |
| Language: | English |
| Additional information: | This version is the author-accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | Machine Learning, Double Descent |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science |
| URI: | https://discovery-pp.ucl.ac.uk/id/eprint/10182015 |
Downloads since deposit
990Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months
Archive Staff Only
 |
View Item |
