Schema Networks : Generalised Policies with Deep Learning
Schema Networks : Generalised Policies with Deep Learning
复制标题
模式网络：深度学习的通用策略
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
Lexing Xie
中科院分区：
文献类型：
--
作者：
S. Toyer;Felipe W. Trevizan;S. Thiébaux;Lexing Xie
在本文中，我们介绍了行动模式网络（ASNet）：一种用于学习概率规划问题的广义策略的神经网络体系结构。通过模仿规划问题的关系结构，ASNets能够采用权重共享方案，该方案允许网络应用于给定规划域中的任何问题。这允许将训练网络的成本分摊到该领域的所有问题上。此外，我们提出了一种训练方法，该方法在小问题上平衡探索和监督训练，以产生在更大问题上评估时保持稳健的策略。在实验中，我们表明ASNet的学习能力使其在几个具有挑战性的领域中显著优于传统的非学习计划器。
In this paper, we introduce the Action Schema Network (ASNet): a neural network architecture for learning generalised policies for probabilistic planning problems. By mimicking the relational structure of planning problems, ASNets are able to adopt a weight sharing scheme which allows the network to be applied to any problem from a given planning domain. This allows the cost of training the network to be amortised over all problems in that domain. Further, we propose a training method which balances exploration and supervised training on small problems to produce a policy which remains robust when evaluated on larger problems. In experiments, we show that ASNet’s learning capability allows it to significantly outperform traditional non-learning planners in several challenging domains.