arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1904.10079cs.LGcs.AIstat.ML
收藏

MineRL 2019年使用人类先验样本高效强化学习竞赛

The MineRL 2019 Competition on Sample Efficient Reinforcement Learning using Human Priors

Carnegie Mellon University(卡内基梅隆大学);Microsoft Research(微软研究院);University of Maryland(马里兰大学);AICrowd;Queen Mary University of London(伦敦玛丽女王大学)
William H. Guss, Cayden Codel, Katja Hofmann, Brandon Houghton, Noboru Kuno, Stephanie Milani, Sharada Mohanty, Diego Perez Liebana, Ruslan Salakhutdinov, Nicho… 展开作者

William H. Guss, Cayden Codel, Katja Hofmann, Brandon Houghton, Noboru Kuno, Stephanie Milani, Sharada Mohanty, Diego Perez Liebana, Ruslan Salakhutdinov, Nicholay Topin, Manuela Veloso, Phillip Wang

摘要:

虽然深度强化学习在许多困难领域取得了突破,但这些成功需要越来越多的样本。由于最先进的强化学习(RL)系统需要呈指数级增长的样本数量,因此它们的开发仅限于人工智能社区中不断缩小的部分。同样,这些系统中的许多不能应用于现实世界的问题,其中环境样本是昂贵的。解决这些限制需要新的,样品效率高的方法。为了促进这一方向的研究,我们引入了MineRL竞赛,使用人类先验知识进行样本有效的强化学习。比赛的主要目标是促进算法的开发,这些算法可以有效地利用人类演示,从而大大减少解决复杂,分层和稀疏环境所需的样本数量。为此,我们介绍:(1)Minecraft ObtainDiamond任务,一个需要长期规划,分层控制和有效探索方法的顺序决策环境;(2)MineRL-v0数据集,一个超过6000万对人类演示的状态-动作对的大规模集合,可以重新模拟成具体的轨迹,并对游戏状态和视觉效果进行任意修改。参赛者将竞争开发系统,解决从环境模拟器,马尔默有限数量的样本获得钻石的任务。比赛分为两轮,参赛者将获得几个配对版本的数据集和具有不同游戏纹理的环境。在每轮比赛结束时,参赛者将提交其学习算法的容器化版本,然后他们将在预先指定的硬件平台上,在一个保持不变的磁盘环境对上从头开始进行总共4天的训练/评估。

英文摘要:

Though deep reinforcement learning has led to breakthroughs in many difficult domains, these successes have required an ever-increasing number of samples. As state-of-the-art reinforcement learning (RL) systems require an exponentially increasing number of samples, their development is restricted to a continually shrinking segment of the AI community. Likewise, many of these systems cannot be applied to real-world problems, where environment samples are expensive. Resolution of these limitations requires new, sample-efficient methods. To facilitate research in this direction, we introduce the MineRL Competition on Sample Efficient Reinforcement Learning using Human Priors. The primary goal of the competition is to foster the development of algorithms which can efficiently leverage human demonstrations to drastically reduce the number of samples needed to solve complex, hierarchical, and sparse environments. To that end, we introduce: (1) the Minecraft ObtainDiamond task, a sequential decision making environment requiring long-term planning, hierarchical control, and efficient exploration methods; and (2) the MineRL-v0 dataset, a large-scale collection of over 60 million state-action pairs of human demonstrations that can be resimulated into embodied trajectories with arbitrary modifications to game state and visuals. Participants will compete to develop systems which solve the ObtainDiamond task with a limited number of samples from the environment simulator, Malmo. The competition is structured into two rounds in which competitors are provided several paired versions of the dataset and environment with different game textures. At the end of each round, competitors will submit containerized versions of their learning algorithms and they will then be trained/evaluated from scratch on a hold-out dataset-environment pair for a total of 4-days on a prespecified hardware platform.

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。

↑