CAREER: Supervised Learning for Incomplete and Uncertain Data

职业：不完整和不确定数据的监督学习

• 批准号: 1350078
• 负责人: Alina Zare
• 金额: $ 45.41万
• 依托单位: University of Missouri-Columbia
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-05-01 至 2017-04-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1350078&HistoricalAwards=false
• 关键词: CAREER; Supervised; Learning; Incomplete; Uncertain

This CAREER project will advance the state of the art in supervised machine learning to allow for incomplete, uncertain and unspecific label information. Supervised machine learning algorithms produce desired outputs for given input data by learning from example training data. The methods generally rely on completely and accurately labeled training data to drive the learning algorithm. However, many applications are plagued with labels that are incomplete, uncertain, and unspecific (lack precision). Current techniques do not adequately handle such data.For example, analysis of satellite imagery to identify the content of each pixel is often conducted by coupling unsupervised learning methods (that do not rely on labeled training data) with manual exploration. This is time-consuming, error-prone, and expensive. Imagine, instead, easy-to-use tools that could understand the content of each pixel in satellite imagery. Extremely large amounts of road map data (for example from Google Maps or OpenStreetMap) and social media information (for example geo-tagged photographs, video clips, and social networking posts) are continually collected and stored. These data could be used as sparsely-labeled training data (with varying degrees of specificity and uncertainty) to guide understanding of satellite imagery.Although the data is available, algorithms have yet to be developed to combine these data sources and identify the content of pixels in satellite images. This work will advance this and other potential applications of machine learning where incomplete, uncertain and unspecific labels in training data challenge the development of effective machine learning algorithms. This CAREER project will achieve these advances through the following research objectives:(1) Investigate and develop a mathematical framework and associated algorithms for Multiple Instance Function Learning that addresses linear and non-linear classification and regression problems with varying levels and types of sparsity, uncertainty, and specificity in training labels.(2) Study and apply the proposed framework and algorithms towards the fusion of satellite imagery, road map data and social media for global scene understanding. This research will be conducted in conjunction with integrated education and outreach activities. In particular, an interactive web application will be developed to provide an avenue for introducing concepts from machine learning and remote sensing to the public for dissemination and outreach. This interactive web application will also be used, along with additional hands-on activities, to introduce high school students to machine learning and remote sensing concepts during an annual summer engineering camp held at the University of Missouri in Columbia, MO. Paired with the web application will be a research website in which data, code, publications and presentations will be shared with the research community. Furthermore, undergraduate and graduate research assistants will be trained in the areas of machine learning and remote sensing. Finally, relevant research topics will be introduced in the PI's undergraduate and graduate courses.

这个CAREER项目将推进监督机器学习的最新技术，以允许不完整、不确定和不特定的标签信息。监督机器学习算法通过从示例训练数据中学习，为给定的输入数据产生期望的输出。这些方法通常依赖于完整和准确标记的训练数据来驱动学习算法。然而，许多应用程序都受到标签不完整、不确定和不具体（缺乏精度）的困扰。目前的技术不能充分处理这类数据。例如，对卫星图像进行分析以识别每个像素的内容通常是通过将无监督学习方法（不依赖于标记的训练数据）与人工探索相结合来进行的。这既耗时又容易出错，而且代价高昂。想象一下，可以理解卫星图像中每个像素的内容的易于使用的工具。大量的路线图数据（例如来自谷歌Maps或OpenStreetMap）和社交媒体信息（例如地理标记的照片、视频剪辑和社交网络帖子）被不断收集和存储。这些数据可以作为稀疏标记的训练数据（具有不同程度的特异性和不确定性）来指导对卫星图像的理解。虽然数据是可用的，但还没有开发算法来结合这些数据源并识别卫星图像中像素的内容。这项工作将推动机器学习的其他潜在应用，其中训练数据中的不完整，不确定和非特定标签挑战了有效机器学习算法的发展。本CAREER项目将通过以下研究目标实现这些进步：(1)研究和开发多实例函数学习的数学框架和相关算法，以解决训练标签中具有不同级别和类型的稀疏性、不确定性和特异性的线性和非线性分类和回归问题。(2)研究并应用所提出的框架和算法，实现卫星图像、路线图数据和社交媒体的融合，实现全球场景理解。这项研究将与综合教育和外联活动一起进行。特别是，将开发一个交互式网络应用程序，为向公众介绍机器学习和遥感的概念提供途径，以便传播和推广。在密苏里州哥伦比亚市密苏里大学举行的一年一度的夏季工程夏令营中，这个交互式网络应用程序还将与其他实践活动一起用于向高中生介绍机器学习和遥感概念。与该网络应用程序相结合的是一个研究网站，其中的数据、代码、出版物和演示文稿将与研究社区共享。此外，本科生和研究生的研究助理将接受机器学习和遥感领域的培训。最后，在PI的本科和研究生课程中介绍相关的研究课题。