Timeline for What is wrong with a neural network model which is so dependent on the seed of initialization?
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 14, 2020 at 20:01 | answer | added | 10xAI | timeline score: 0 | |
| May 14, 2020 at 17:12 | comment | added | user137927 | I changed the learning rate a lot. I think the problem is that there are a lot of local optima in the loss function. I don't know whether I should reduce the dimension of the input features with the PCA or I should focus on changing the loss function or other solutions. | |
| May 14, 2020 at 14:58 | answer | added | Tinu | timeline score: 1 | |
| May 14, 2020 at 14:53 | comment | added | Sean Payne | It's hard to know without seeing your model, but an off the cuff guess would be that your learning rate may be off and the better test results were a lucky find of the gradient. | |
| May 14, 2020 at 14:31 | history | asked | user137927 | CC BY-SA 4.0 |