arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1902.01569cs.AIcs.RO
收藏

在模拟中学习

Learning to Learn in Simulation

Carnegie Mellon University(卡内基梅隆大学);Moffett Field, CA(莫菲特菲尔德,加利福尼亚州)

Ervin Teng, Bob Iannucci

摘要:

深度学习通常需要手动收集和注释训练集。在机器人平台上,我们能否通过训练机器人保持好奇心来部分自动化这项任务,即,在环境中寻找有益的培训信息?在这项工作中,我们解决了好奇心的问题,因为它涉及到在线,实时,人在回路中的无人机上的目标检测算法的训练,其中运动被限制在两个维度。我们使用3D仿真环境和深度强化学习来训练好奇代理,进而训练目标检测模型。这个代理可能有两个相互冲突的目标之一:尽可能快地进行训练,或者以最少的人力投入进行训练。我们概述了一个奖励函数,它允许好奇心代理学习这些目标中的任何一个,同时考虑到它要运行的无人机平台的一些物理特性。此外,我们表明,我们可以通过调整奖励函数中的参数来衡量实现这些目标的重要性。

英文摘要:

Deep learning often requires the manual collection and annotation of a training set. On robotic platforms, can we partially automate this task by training the robot to be curious, i.e., to seek out beneficial training information in the environment? In this work, we address the problem of curiosity as it relates to online, real-time, human-in-the-loop training of an object detection algorithm onboard a drone, where motion is constrained to two dimensions. We use a 3D simulation environment and deep reinforcement learning to train a curiosity agent to, in turn, train the object detection model. This agent could have one of two conflicting objectives: train as quickly as possible, or train with minimal human input. We outline a reward function that allows the curiosity agent to learn either of these objectives, while taking into account some of the physical characteristics of the drone platform on which it is meant to run. In addition, We show that we can weigh the importance of achieving these objectives by adjusting a parameter in the reward function.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑