课题基金 / 基金详情

RI: Small: Reinforcement Learning by Mirror Descent

RI: Small: Reinforcement Learning by Mirror Descent
RI:小:通过镜像下降的强化学习
批准号:
1216467
负责人:
Sridhar Mahadevan
金额:
$45.0万
依托单位国家:
美国
项目类别:
Standard Grant
财政年份:
2012
资助国家:
美国
项目状态:
已结题
起止时间:
2012-08-01 至 2016-07-31

项目摘要

项目成果

Sridhar Mahadevan的其他基金

相似基金

相关文献

中文摘要
翻译
机器学习的一个基本挑战是计算代理的设计,而不是被明确编程,在随机的现实世界环境中自主学习复杂的任务。过去的方法,如用于解决马尔可夫决策过程的强化学习算法,在大型状态空间上的扩展性很差。提出的研究通过研究一种结合强化学习和在线凸优化的新框架,特别是镜像下降和相关算法,解决了这种维度诅咒。通过使用特定于特定状态空间几何的距离生成函数,镜像下降在高维状态空间中的尺度明显优于经典的一阶梯度下降。提出的框架在自主机器学习代理的设计中实现了几个重要的算法进步:将开发一类新的基于一阶镜像下降的方法,用于学习马尔可夫决策过程的稀疏解,其规模明显优于以前的二阶方法;研究求解半马尔可夫决策过程的新层次方法;最后,将探讨在各种高维马尔可夫决策过程中的应用。所提出的工作的预期结果包括在设计学习解决顺序决策问题的自主代理方面取得基础进展,这将影响从制造到机器人和调度的大量目标应用。教育目标包括开发用于顺序决策的在线凸优化的研究生水平课程,以及跨学科教程,以加强从应用数学和优化到机器学习和人工智能的思想的交叉施肥。
英文摘要
A fundamental challenge in machine learning is the design of computational agents that, rather than being explicitly programmed, autonomously learn complex tasks in stochastic real-world environments. Past approaches, such as reinforcement learning algorithms for solving Markov decision processes, scale poorly to large state spaces. The proposed research addresses this curse of dimensionality by investigating a novel framework combining reinforcement learning and online convex optimization, in particular mirror descent and related algorithms. Mirror descent scales significantly better than classical first-order gradient descent in high-dimensional state spaces, by using a distance-generating function specific to a particular state space geometry.The proposed framework enables several significant algorithmic advances in the design of autonomous machine learning agents: a new class of first-order mirror-descent based methods for learning sparse solutions to Markov decision processes will be developed that scale significantly significantly better than previous second-order methods; novel hierarchical methods for solving semi-Markov decision processes will be investigated; and finally, applications to a variety of high-dimensional Markov decision processes will be explored.The anticipated outcomes of the proposed work include foundational advances in designing autonomous agents that learn to solve sequential decision-making problems, which will impact a large number of target applications from manufacturing to robotics and scheduling. The educational goal includes the development of a graduate-level course in online convex optimization for sequential decision-making, as well as interdisciplinary tutorials to enhance the cross-fertilization of ideas from applied mathematics and optimization to machine learning and artificial intelligence.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文
Collaborative Research: Transfer Learning for Chemical Analyses from Laser-Induced Spectroscopy
  • 批准号:
    1307179
  • 项目类别:
    Standard Grant
  • 资助金额:
    $15.89万
  • 财政年份:
    2013
  • 负责人:
    Sridhar Mahadevan
  • 依托单位:
NeTS Small: Analysis and Design of Best-Effort Content-Caching Networks
  • 批准号:
    1117764
  • 项目类别:
    Standard Grant
  • 资助金额:
    $30.0万
  • 财政年份:
    2011
  • 负责人:
    Sridhar Mahadevan
  • 依托单位:
Manifold Alignment of High-Dimensional Data Sets
  • 批准号:
    1025120
  • 项目类别:
    Standard Grant
  • 资助金额:
    $49.99万
  • 财政年份:
    2010
  • 负责人:
    Sridhar Mahadevan
  • 依托单位:
RI-Medium: Collaborative Research: Learning Multiscale Representations using Harmonic Analysis on Graphs
  • 批准号:
    0803288
  • 项目类别:
    Standard Grant
  • 资助金额:
    $34.52万
  • 财政年份:
    2008
  • 负责人:
    Sridhar Mahadevan
  • 依托单位:
国内基金
海外基金
昼夜节律性small RNA在血斑形成时间推断中的法医学应用研究
  • 批准号:
  • 项目类别:
    省市级项目
  • 资助金额:
    --
  • 批准年份:
    2024
  • 负责人:
  • 依托单位:
tRNA-derived small RNA上调YBX1/CCL5通路参与硼替佐米诱导慢性疼痛的机制研究
  • 批准号:
  • 项目类别:
    省市级项目
  • 资助金额:
    10.0万元
  • 批准年份:
    2022
  • 负责人:
    张祥忠
  • 依托单位:
Small RNA调控I-F型CRISPR-Cas适应性免疫性的应答及分子机制
Small RNAs调控解淀粉芽胞杆菌FZB42生防功能的机制研究
  • 批准号:
    31972324
  • 项目类别:
    面上项目
  • 资助金额:
    58.0万元
  • 批准年份:
    2019
  • 负责人:
    高学文
  • 依托单位: