Timeline for Should I be decaying the learning rate and the exploration rate in the same manner?
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 7, 2022 at 16:15 | history | edited | nbro | CC BY-SA 4.0 |
edited tags; edited title
|
| Sep 12, 2018 at 8:07 | comment | added | 16Aghnar | So, a lower LR means a slower convergence but an improved asymptote (limit of the learning curve). So tuning it depends on the time you have, and also on your model. You can begin with, for example, 0.001, see the learning curve, and if you reach quickly the asymptote, you can try with a lower LR, see again the learning curve, and so on. (and notice that the 0.98 - 0.997 values I mentioned in my answer are for the LR decay, not for the LR) | |
| Sep 12, 2018 at 0:12 | comment | added | rtz | @16Aghnar in what scenario would I use a lower learning rate? I've looked at the ATARI papers and games like Super Mario Bros. They used learning rates of 0.00025. Is it because we want it to not get hooked onto the same decision again and again? As in, with a higher learning rate, it would assume action x would be best. I hope I made sense! | |
| Sep 11, 2018 at 22:50 | vote | accept | rtz | ||
| Sep 11, 2018 at 11:13 | answer | added | 16Aghnar | timeline score: 6 | |
| Sep 11, 2018 at 10:55 | review | First posts | |||
| Sep 11, 2018 at 11:38 | |||||
| Sep 11, 2018 at 10:47 | history | asked | rtz | CC BY-SA 4.0 |