4 ms·
You don't need to update any weights, the model is already trained.
by drexlspivey 8y ago
You don't need to update any weights, the model is already trained.
- minimaxir 8y agoMy initial read of https://blog.openai.com/openai-five/ https://blog.openai.com/openai-five/ was that the 1024-cell LSTM game state layer was trained on the fly, but after rereading it may not be the case.
- lma21 8y agoIgnorant here. How do you take an already trained model and execute it elsewhere? Isn’t the training phase part of the whole (ongoing) simulation?
- sethgecko 8y agoThe training phase is when you are trying to minimise your loss function by trying different weights each iteration. Once that is finished, you have a trained model which you can use by providing input and getting an output (with the weights frozen). The training phase is very expensive computationally because you have to calculate the gradient of your loss function on potentially huge tensors. The execution phase is not that expensive and commercial laptops will be likely able to run the model without any problems.