RT @OriolVinyalsML: Great SIGGRAPH paper which uses RL/ML for motion generation. I predict many more applications of machine learning to gr…
The "Supplementary Materials" for DeepMimic actually give the clearest, more concise explanation there is of n-step return, λ-return, TD(λ) and GAE(λ) : https://t.co/ceCgsRVNPp https://t.co/uP2phtYTe4
@GaryMarcus Best from.. DL world =  ‘physics prior’ *learned* & then applied with in RL scheme. CG world =  ‘physics prior’ encoded as heuristic for specific object & then used in RL scheme w/ goal of realistic behavior.  https://t.
siggraph papers that stood out: deepmimic, example guided deep reinforcement learning of physical based character skills, xue bin peng, pieter eabbeel, sergey levine, michiel van de panne (of @ubccs!) https://t.co/gC4yAqqmJF #ieeevis #vds