Some Works abount RL and Meta-Learning in ICML 2019 (Part2)

Making Deep Q-learning methods robust to time discretization [Website] Corentin Tallec · Leonard Blier · Yann Ollivier Universit´e Paris-Sud, Facebook AI Research Notes Problem: Deep Reinforcement Learning is not robust to hyperparameterization, implementation details, or small environment changes. In this paper, we identify sensitivity to time discretization in near continuous-time... [Read More]

Framework of Pick and Place for Mobile Manipulation

The overall approach to the pick and place task is shown schematically in Figure 1. A representation of the environment is built and maintained using a combination of 2D and 3D visual sensors. This model includes a lower resolution representation of the environment using an octree-based approach, and higher resolution... [Read More]
Tags: robotics

Planning and Control in Self-Driving Cars

Illustration of the hierarchy of decision-making processes. A destination is passed to a route planner that generates a route through the road network. A behavioral layer reasons about the environment and generates a motion specification to progress along the selected route. A motion planner then solves for a feasible motion... [Read More]


今天第一次使用Markdown,被它的效果深深地感动了。之前电脑安装了Markdown编辑器 — MarkdownPad2,但因为种种原因没有去使用。今天在看魏剑锋的微信公众号文章时,看到了几篇关于写作相关的文章,就想到了之前下载MarkdownPad2就是因为想写博客,于是就去尝试着用了一下。 [Read More]
Tags: writing