Timeline for Generative models for time series simulation
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 19, 2017 at 18:16 | comment | added | DiveIntoML | @gnjago Any supervised generative model will do. The key here is to use supervised model instead of unsupervised GAN. You can simply regard whatever you want to generate as y and whatever you want to condition upon as x, no matter if x is single step or multistep. | |
| Oct 19, 2017 at 18:14 | history | edited | DiveIntoML | CC BY-SA 3.0 |
added 13 characters in body
|
| Oct 18, 2017 at 8:02 | comment | added | gnjago | What to do if we need to look at multiple timesteps before predicting nexttime step. Is LSTM the only option? I am not looking to predict the next time step. Just looking to generate data with multiple time series in it. Will temporal probabilistic graphical models be good? | |
| Aug 18, 2017 at 2:24 | review | Late answers | |||
| Aug 18, 2017 at 2:37 | |||||
| Aug 18, 2017 at 2:08 | history | answered | DiveIntoML | CC BY-SA 3.0 |