ha, nice. thought evo and RL were same same by 'generate and test', e.g. similarity to CACLA
"Evolution Strategies as a Scalable Alternative to RL" --looks like a very nice, forward-looking paper from OpenAI.https://arxiv.org/abs/1703.03864
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
so randomly guessing is the new state of the art?
-
Yes, Evolution strategies are also better than reinforcement learning for molecule generationhttps://medium.com/the-ai-lab/you-didnt-need-deep-learning-to-generate-new-molecules-4c784747b2cc …
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.