qq_33331451的博客

Deep Reinforcement Learning for Unsupervised Video Summarization with Diversity-Representativeness Rewardpaper: https://arxiv.org/abs/1801.00054code: https://github.com/KaiyangZhou/pytorch-vsumm-reinforceconference: AAAI 2018Abstract  视频摘要的目的是为了方便大规.

Continuous control with deep reinforcement learning(ICLR2016)https://arxiv.org/abs/1509.02971v2文章目录AbstractIntroductionAbstract   作者将Deep-Learning成功的应用于连续动作领域。作者提出了一个基于确定性策略梯度(deterministic policy gradient)的无模型(model-free)的角色批判(actor-critic)算法，它可以在连续.