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 1904.02206cs.LGstat.ML
收藏

深度强化学习的联合预训练与监督,自动编码器和值损失

Jointly Pre-training with Supervised, Autoencoder, and Value Losses for Deep Reinforcement Learning

Washington State University(华盛顿州立大学)

Gabriel V. de la Cruz, Yunshu Du, Matthew E. Taylor

摘要:

深度强化学习(DRL)算法被认为是数据低效的。一个原因是DRL代理学习功能和策略表。将先验知识集成到DRL算法中是提高学习效率的一种方法,因为它有助于构建有用的表示。在这项工作中,我们考虑将人类知识,以加速异步优势演员-评论家(A3 C)算法通过预训练少量的非专家的人类示范。我们利用监督自动编码器框架,并提出了一种新的预训练策略,该策略联合训练加权监督分类损失,无监督重建损失和预期返回损失。与以监督或无监督方式独立训练相比,所得到的预训练模型学习到更多有用的特征。我们的预训练方法大大提高了A3 C代理在Atari游戏Pong和MsPacman中的学习性能,在更少的游戏交互次数下超过了最先进算法的性能。我们的方法是轻量级的,易于实现在一个单一的机器。为了重现性,我们的代码可在github.com/gabrieledcjr/DeepRL/tree/A3C-ALA2019上获得

英文摘要:

Deep Reinforcement Learning (DRL) algorithms are known to be data inefficient. One reason is that a DRL agent learns both the feature and the policy tabula rasa. Integrating prior knowledge into DRL algorithms is one way to improve learning efficiency since it helps to build helpful representations. In this work, we consider incorporating human knowledge to accelerate the asynchronous advantage actor-critic (A3C) algorithm by pre-training a small amount of non-expert human demonstrations. We leverage the supervised autoencoder framework and propose a novel pre-training strategy that jointly trains a weighted supervised classification loss, an unsupervised reconstruction loss, and an expected return loss. The resulting pre-trained model learns more useful features compared to independently training in supervised or unsupervised fashion. Our pre-training method drastically improved the learning performance of the A3C agent in Atari games of Pong and MsPacman, exceeding the performance of the state-of-the-art algorithms at a much smaller number of game interactions. Our method is light-weight and easy to implement in a single machine. For reproducibility, our code is available at github.com/gabrieledcjr/DeepRL/tree/A3C-ALA2019

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。

↑