arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1401.3434cs.LG
收藏

自适应随机资源控制:一种机器学习方法

Adaptive Stochastic Resource Control: A Machine Learning Approach

Computer and Automation Research Institute, Hungarian Academy of Sciences(匈牙利科学院计算机与自动化研究所);Faculty of Mechanical Engineering, Budapest University of Technology and Economics(布达佩斯技术与经济大学机械工程学院)

Balázs Csanád Csáji, László Monostori

摘要:

本文研究了一类随机资源分配问题,该问题具有稀缺的、可重用的资源和非抢占的、时间依赖的、相互关联的任务。这种方法是几个标准的资源管理问题,如调度和运输问题的自然概括。首先,反应性的解决方案被认为是和定义为适当重新制定的马尔可夫决策过程(MDP)的控制策略。我们认为,这种重新制定有几个有利的性质,如它有有限的状态和行动空间,它是非周期的,因此所有的政策是适当的,控制政策的空间可以安全地限制。接下来,近似动态规划(ADP)方法,如拟合Q学习,建议计算一个有效的控制策略。为了实现对剩余成本函数的复杂维护,研究了两种表示方法:哈希表和支持向量回归(SVR),特别是nu-SVR。一些额外的改进,如有限前瞻推出算法在初始阶段的应用,动作空间分解,任务聚类和分布式采样进行了研究。最后,基准和行业相关数据的实验结果。

英文摘要:

The paper investigates stochastic resource allocation problems with scarce, reusable resources and non-preemtive, time-dependent, interconnected tasks. This approach is a natural generalization of several standard resource management problems, such as scheduling and transportation problems. First, reactive solutions are considered and defined as control policies of suitably reformulated Markov decision processes (MDPs). We argue that this reformulation has several favorable properties, such as it has finite state and action spaces, it is aperiodic, hence all policies are proper and the space of control policies can be safely restricted. Next, approximate dynamic programming (ADP) methods, such as fitted Q-learning, are suggested for computing an efficient control policy. In order to compactly maintain the cost-to-go function, two representations are studied: hash tables and support vector regression (SVR), particularly, nu-SVRs. Several additional improvements, such as the application of limited-lookahead rollout algorithms in the initial phases, action space decomposition, task clustering and distributed sampling are investigated, too. Finally, experimental results on both benchmark and industry-related data are presented.

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。

↑