Timeline for When training a neural network, why choose Adam over L-BGFS for the optimizer?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 13, 2020 at 18:55 | comment | added | GoHokies | you're welcome! here's a nice visual explanation of momentum: distill.pub/2017/momentum | |
| Jan 13, 2020 at 16:50 | comment | added | tietäjä | Excellent! Thank you. I had come across the first StackExchange post and was pretty disappointed with most of the replies. But I hadn't discovered the other resources you shared. Seems they outline some significant challenges for second-order methods in the realm of machine learning. And that explanation of the concept of "momentum" in Adam was brilliant. You've given me a lot to dig into. | |
| Jan 13, 2020 at 16:34 | vote | accept | tietäjä | ||
| Jan 13, 2020 at 12:57 | history | edited | GoHokies | CC BY-SA 4.0 |
added 176 characters in body
|
| Jan 13, 2020 at 10:12 | history | answered | GoHokies | CC BY-SA 4.0 |