课题基金 / 基金详情

RI: Medium: Advances and Applications in Submodularity for Machine Learning

RI: Medium: Advances and Applications in Submodularity for Machine Learning
RI:媒介:机器学习子模块性的进展和应用
批准号:
1162606
负责人:
Jeffrey Bilmes
金额:
$81.45万
依托单位:
依托单位国家:
美国
项目类别:
Continuing Grant
财政年份:
2012
资助国家:
美国
项目状态:
已结题
起止时间:
2012-07-01 至 2018-06-30

项目摘要

项目成果

Jeffrey Bilmes的其他基金

相似基金

相关文献

中文摘要
翻译
子模块性是一种直观的收益递减性质,说明将一个元素添加到一个较小的集合中比将它添加到一个较大的集合中更有帮助。子模块化允许人们有效地找到离散问题的可证明的最优或近最优解。次模最小化已经在图形模型推理和聚类中得到了应用,而最大化已经被应用于变量/特征选择和主动学习中。然而,子模块化在机器学习及其应用中的适用性才刚刚开始。此外,组合学和运筹学文献中关于子模块优化的工作基本上没有意识到机器学习中出现的独特问题。因此,现有的标准算法没有利用机器学习中出现的子模块问题的某些结构或变体。因此,研究涉及子模块目标的新型机器学习问题可以在纯组合学文献中取得进展。我们建议开展活动,将机器学习和组合优化的研究结合起来,以解决两个社区都无法单独解决的问题。特别是,我们建议使用机器学习的见解来扩大典型的子模块优化问题的规模(通过关注学习中出现的问题实例)。我们还建议进一步绘制子模块化在机器学习中发挥作用的领域。在本次授权中,我们将介绍专门与子模块化相关的新子模块结构。我们将介绍机器学习的子模块学习问题。我们将引入新的带约束的子模优化问题。最后,我们将把这些子模块实例应用到计算机视觉、语音识别和自然语言处理等实际应用中。
英文摘要
Submodularity is an intuitive diminishing returns property, stating that adding an element to a smaller set helps more than adding it to a larger set. Submodularity allows one to efficiently find provably optimal or near-optimal solutions to discrete problems. Submodular minimization has found use, e.g., in graphical model inference and clustering, whereas maximization has been applied, e.g., to variable/feature selection and active learning. Submodularity, however, is still only beginning to show applicability in machine learning and its applications. Moreover, work on submodular optimization in the combinatorics and operations research literature has been primarily unaware of unique problems arising in machine learning. Therefore, existing standard algorithms do not exploit certain structures or variants of the submodular problems arising in machine learning. Studying novel machine learning problems involving submodular objectives can thus lead to advances in the pure combinatorics literature. We propose to pursue activities that bring together research in machine learning and combinatorial optimization to solve problems which neither of the communities can solve alone.In particular, we propose to use insights from machine learning to enable scaling up typical submodular optimization problem sizes (by focusing on problem instances arising in learning). We also propose to further chart the territory that submodularity plays in machine learning. In this grant, we will introduce new submodular structures specifically related to submodularity. We will introduce submodular learning problems for machine learning. We will introduce new submodular optimization problems with constraints. And lastly, we will apply these submodular instances to real-world applications in computer vision, speech recognition, and natural language processing.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文
Collaborative Research: RI: Medium: Submodular Information Functions with Applications to Machine Learning
  • 批准号:
    2106389
  • 项目类别:
    Standard Grant
  • 资助金额:
    $60.0万
  • 财政年份:
    2021
  • 负责人:
    Jeffrey Bilmes
  • 依托单位:
CI-ADDO-EN: Software Infrastructure for Temporal Modeling
  • 批准号:
    0855230
  • 项目类别:
    Continuing Grant
  • 资助金额:
    $69.13万
  • 财政年份:
    2009
  • 负责人:
    Jeffrey Bilmes
  • 依托单位:
RI: Medium: Collaborative Research: Explicit Articulatory Models of Spoken Language, with Application to Automatic Speech Recognition
  • 批准号:
    0905341
  • 项目类别:
    Standard Grant
  • 资助金额:
    $37.8万
  • 财政年份:
    2009
  • 负责人:
    Jeffrey Bilmes
  • 依托单位:
Intransitive Classification and Choice
  • 批准号:
    0535100
  • 项目类别:
    Continuing Grant
  • 资助金额:
    $48.02万
  • 财政年份:
    2005
  • 负责人:
    Jeffrey Bilmes
  • 依托单位:
海外基金