摘要
一位新闻记者-机器人与机器学习的工作人员新闻编辑每日新闻-交通研究发现-智能车辆在一份新的报告中讨论。根据NewsRx记者在意大利佛罗伦萨的新闻报道，研究表明，“条件模拟学习是训练自动驾驶AGEN TS的一种常见和有效的方法。然而，有两个问题限制了这种方法的全部潜力：(i)惰性IA问题，这是一种因果混淆的特殊情况，在这种情况下，代理人错误地将低速与没有加速联系起来。”以及(ii)离线和在线性能之间的相关性较低，这是由于小错误的累积导致年龄nt处于以前看不见的状态。"
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News – Research findings on Transportation - Intelligent Vehicles are discussed in a new report. According to news reporting originating in Florence, Italy, by NewsRx journalists, research stated, “Conditional Imitat ion learning is a common and effective approach to train autonomous driving agen ts. However, two issues limit the full potential of this approach: (i) the inert ia problem, a special case of causal confusion where the agent mistakenly correl ates low speed with no acceleration, and (ii) low correlation between offline an d online performance due to the accumulation of small errors that brings the age nt in a previously unseen state.”
NETL
NSTL