Now also on ArXiv: "ALLSTEPS: Curriculum-driven Learning of Stepping Stone Skills"https://arxiv.org/abs/2005.04323
-
-
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I've completed obbys 3 times harder in Roblox.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
What's the sensor the agent is using here?
-
From skimming the arxiv paper, a ~60 dimensional state vector.
- Show replies
New conversation -
-
-
Awesome work!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thank you!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Ever try binarizing the policy output? bang-bang seems to emerge anyway a lot of the time so I'm curious if it would hamper the agent here or if its really making use of the action space.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Impressive work by our colleagues at University of Columbia (Zhaoming Xie Hung Yu Ling Nam Hee Kim Michiel van de Panne) Congrats !
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This looks great!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This is great.
@punchesbears you might be interested.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.