Vic_Hao的博客

这篇文章的目标是将深度神经网络模型在其他领域中的成功扩展到基于模型的强化学习中。The contribution of this paper is:They demonstrate effective model-based reinforcement learning with neural network models for several contact-rich simulat...

Research Topic得到一个model, 输入是raw high-dimensional sensory input，输出是Q value functionReinforcement learning presents several challenges from a deep learning perspective:most successful deep learning ...