The parameters θ of the moth policy and the refined policy.
收藏数据链接：
官方服务：
资源简介：
OSD is the obstacle spatial density feature; OSD(0) and OSD(0.5) denote OSD features with thresholds 0 and 0.5, respectively. OF is the optical flow feature. Percentages are the absolute value of weights normalized by the ℓ1 norm of the weight vector. Reward is the expected average reward (3) for the regressed policy and the optimized policy.
创建时间：
2020-01-09
