 |
ML Newsnew | past | best | rss | submit | about |
| Backprop Evolution (arxiv.org) | |
| 3 points by greenburg on Aug 9, 2018 | hide | past | pdf | discuss on HN | |
|
In plain words: An evolutionary search combines simple math pieces into new versions of the backpropagation rule that adjusts a network's weights, keeping the ones that generalize best after a few training epochs. Several discovered rules learn faster in short runs and match standard backpropagation once trained. AbstractThe back-propagation algorithm is the cornerstone of deep learning. Despite its importance, few variations of the algorithm have been attempted. This work presents an approach to discover new variations of the back-propagation equation. We use a domain specific lan- guage to describe update equations as a list of primitive functions. An evolution-based method is used to discover new propagation rules that maximize the generalization per- formance after a few epochs of training. We find several update equations that can train faster with short training times than standard back-propagation, and perform similar as standard back-propagation at convergence. | |
| add comment on HN |