Parallel Policy-Gradient Methods for Parameter Optimization of Nonlinear Feedback Controllers figure
AlphaXiv 中文概览(可滚动查看)