Abstract
We present an analytic solution to the problem of on-line gradient-descent learning for two-layer neural networks with an arbitrary number of hidden units in both teacher and student networks.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 4337-4340 |
| Number of pages | 4 |
| Journal | Physical review letters |
| Volume | 74 |
| Issue number | 21 |
| DOIs | |
| State | Published - 1995 |
ASJC Scopus subject areas
- General Physics and Astronomy