课题基金 / 基金详情

Linking metric and symbolic levels in autonomous reinforcement learning

Linking metric and symbolic levels in autonomous reinforcement learning
连接自主强化学习中的度量和符号级别
批准号:
200282059
负责人:
Professor Dr. Klaus Obermayer
金额:
$0.0万
依托单位国家:
德国
项目类别:
Priority Programmes
财政年份:
2011
资助国家:
德国
项目状态:
已结题
起止时间:
2010-12-31 至 2019-12-31

项目摘要

项目成果

Professor Dr. Klaus Obermayer的其他基金

相似基金

相关文献

中文摘要
翻译
强化学习(RL)已经成为自主智能体的理论基础,但在实际应用中却很少。一个重要的缺陷是维度的诅咒,我们在之前的SPP-1527资助的项目“大型分解状态空间中的值表示”中解决了这个问题。然而,在我们的工作中,我们发现缺乏对新情况的适应性同样是实际应用的障碍。我们的框架包含一个符号层,借用自关系RL,以使度量转换模型适应新的情况。我们假设RL的度量层和符号层可以互补地工作,以提供许多实际应用中所需的精度和灵活性。为了研究我们开发的框架的可能性,我们正在追求以下科学目标:(1)开发一种贝叶斯方法来积极学习具有关系灵活性的指标过渡模型,以及(2)调查指标和符号规划者层次结构中的协同效应,以改进规划和调整关系符号和行动。设想的方法将把分解的MDP和关系RL的字段转换到一个公共框架中,并允许自上而下和自下而上的适配。我们预计,在强烈依赖于底层度量空间的关系任务中,例如在机器人应用中,会有显著的改进。
英文摘要
Reinforcement learning (RL) has emerged as a well-founded theoretical basis for autonomous agents, but is rarely used in practical applications. One essential flaw is the curse of dimensionality, which we address in our previous SPP-1527 funded project "value representation in large factored state spaces". However, in our work we found the lack of adaptivity to new situations an equally obstructive obstacle for practical applications. Our framework contains a symbolic layer, borrowed from relational RL, to adapt the metric transitions model to new situations. We hypothesize that metric and symbolic layers of RL can work complementary to provide the precision and flexibility required in many practical applications. To investigate the possibilities of our developed framework we are pursuing the following scientific goals: (1) to develop a Bayesian method to actively learn a metric transition model with relational flexibility, and (2) to investigate synergies in the hierarchy of metric and symbolic planners to improve the planning and adjust the relational symbols and actions. The envisioned methods will cast the fields of factored MDP and relational RL into a common framework, and allow top-down and bottom-up adaptation. We expect significant improvement in relational tasks that depend strongly on an underlying metric spaces, for example in robotic applications.
期刊论文(4)
专著(0)
科研奖励(0)
会议论文
DOI: 10.14279/depositonce-5715
发表时间: 2017
期刊:
影响因子: --
作者: [Wendelin Böhmer]
通讯作者: Wendelin Böhmer
A Fenchel-Moreau-Rockafellar type theorem on the Kantorovich-Wasserstein space with applications in partially observable Markov decision processes
Kantorovich-Wasserstein 空间上的 Fenchel-Moreau-Rockafellar 型定理及其在部分可观测马尔可夫决策过程中的应用
DOI: 10.1016/j.jmaa.2019.05.004
发表时间: 2019
期刊: Journal of Mathematical Analysis and Applications
影响因子: 1.3
作者: [Vaios Laschos, Klaus Obermayer, Yun Shen, Wilhelm Stannat]
通讯作者: Wilhelm Stannat
DOI: 10.1007/978-3-319-23528-8_8
发表时间: 2014-12
期刊:
影响因子: --
作者: [Wendelin Böhmer;K. Obermayer]
通讯作者: Wendelin Böhmer;K. Obermayer
DOI: 10.1007/s10994-012-5300-0
发表时间: 2012-06
期刊: Machine Learning
影响因子: 7.5
作者: [Wendelin Böhmer;S. Grünewälder;H. Nickisch;K. Obermayer]
通讯作者: Wendelin Böhmer;S. Grünewälder;H. Nickisch;K. Obermayer
Risk-sensitive choice and reinforcement learning under uncertainty
  • 批准号:
    407012307
  • 项目类别:
    Research Grants
  • 资助金额:
    $0.0万
  • 财政年份:
    2018
  • 负责人:
    Professor Dr. Klaus Obermayer
  • 依托单位:
Risk-sensitive decision making under inclomplete information
Lernende Software-Agenten zur Filterung von Textdokumenten
Neuronale biologisch inspirierte Steuerungsachitektur für einen mobilen Roboter
国内基金
海外基金
基于异构医学影像数据的深度挖掘技术及中枢神经系统重大疾病的精准预测
  • 批准号:
    61672236
  • 项目类别:
    面上项目
  • 资助金额:
    64.0万元
  • 批准年份:
    2016
  • 负责人:
    王骏
  • 依托单位: