Timeline for Rate of convergence for cyclic gradient descent
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 14, 2015 at 3:16 | vote | accept | Janathan | ||
| May 14, 2015 at 3:16 | vote | accept | Janathan | ||
| May 14, 2015 at 3:16 | |||||
| May 14, 2015 at 3:02 | vote | accept | Janathan | ||
| May 14, 2015 at 3:16 | |||||
| May 14, 2015 at 2:59 | comment | added | Janathan | Thanks for the answer. The paper by Nedic and Bertsekas is very relevant. I hoped that the rate could be independent of n, which is the case in SGD. However, this seems hard for incremental subgradient. | |
| May 13, 2015 at 7:00 | history | answered | Dirk | CC BY-SA 3.0 |