Semantic Predictive Control For Explainable And Efficient Policy Learning
2019 INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA)(2019)
摘要
Visual anticipation of ego and object motion over a short time horizons is a key feature of human-level performance in complex environments. We propose a driving policy learning framework that predicts feature representations of future visual inputs; our predictive model infers not only future events but also semantics, which provide a visual explanation of policy decisions. Our Semantic Predictive Control (SPC) framework predicts future semantic segmentation and events by aggregating multi-scale feature maps. A guidance model assists action selection and enables efficient sampling-based optimization. Experiments on multiple simulation environments show that networks which implement SPC can outperform existing model-based reinforcement learning algorithms in terms of data efficiency and total rewards while providing clear explanations for the policy's behavior.
更多查看译文
关键词
visual explanation,policy decisions,SPC,future semantic segmentation,multiscale feature maps,guidance model,multiple simulation environments,model-based reinforcement,data efficiency,short time horizons,human-level performance,complex environments,driving policy learning framework,feature representations,sampling-based optimization,semantic predictive control framework
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络