ขนาดวิดีโอ: 1280 X 720853 X 480640 X 360
แสดงแผงควบคุมโปรแกรมเล่น
เล่นอัตโนมัติ
เล่นใหม่
Great , clear explanation ... Thanks for making the vid !
Thanks for the video...i was working on the car racing experiment......this explanation helped a lot.
Cool! I was thinking for some time to do a channel where I talk about papers like this! Glad to see you actually are doing it, so useful :)!
You can still do yours. I will be expecting
You can still start yours, its never too late to start!
This is pretty much the same idea as the Dreamer paper (which was published later!)
very great
wouldnt using an actor based on reconstructe frame have been better here ?
Question: If the RNN generates a new $z$, why bother feeding it through the decoder (& then back through the encoder) to get the $z$ for the next timestep?th-cam.com/video/dPsXxLyqpfs/w-d-xo.html
Great , clear explanation ... Thanks for making the vid !
Thanks for the video...i was working on the car racing experiment......this explanation helped a lot.
Cool! I was thinking for some time to do a channel where I talk about papers like this! Glad to see you actually are doing it, so useful :)!
You can still do yours. I will be expecting
You can still start yours, its never too late to start!
This is pretty much the same idea as the Dreamer paper (which was published later!)
very great
wouldnt using an actor based on reconstructe frame have been better here ?
Question: If the RNN generates a new $z$, why bother feeding it through the decoder (& then back through the encoder) to get the $z$ for the next timestep?
th-cam.com/video/dPsXxLyqpfs/w-d-xo.html