Material Type  

 학위논문

 

0017161295

Date and Time of Latest Transaction  

20250211151336

ISBN  

9798383162866

DDC  

310

Author  

Yuan, Gan.

Title/Author  

Topics on Machine Learning Under Imperfect Supervision.

Publish Info  

[S.l.] : Columbia University., 2024

Publish Info  

Ann Arbor : ProQuest Dissertations & Theses, 2024

Material Info  

128 p.

General Note  

Source: Dissertations Abstracts International, Volume: 85-12, Section: A.

General Note  

Advisor: Kpotufe, Samory;Zheng, Tian.

학위논문주기  

Thesis (Ph.D.)--Columbia University, 2024.

Abstracts/Etc  

요약This dissertation comprises several studies addressing supervised learning problems where the supervision is imperfect.Firstly, we investigate the margin conditions in active learning. Active learning is characterized by its special mechanism where the learner can sample freely over the feature space and exploit mostly of the limited labelling budget by querying the most informative labels. Our primary focus is to discern critical conditions under which certain active learning algorithms can outperform the optimal passive learning minimax rate. Within a non-parametric multi-class classification framework, our results reveal that the uniqueness of Bayes labels across the feature space serves as the pivotal determinant for the superiority of active learning over passive learning.Secondly, we study the estimation of central mean subspace (CMS), and its application in transfer learning. We show that a fast parametric convergence rate of form Cd · n −1/2 is achievable via estimating the expected smoothed gradient outer product, for a general class of covarite distribution that admits Gaussian or heavier distributions. When the link function is a polynomial with a degree of at most r and the covariates follows the standard Gaussian, we show that the prefactor depends on the ambient dimension d as Cd ∝ d r . Furthermore, we show that under a transfer learning setting, an oracle rate of prediction error as if the CMS is known is achievable, when the source training data is abundant.Finally, we present an innovative application involving the utilization of weak (noisy) labels for addressing an Individual Tree Crown (ITC) segmentation challenge. Here, the objective is to delineate individual tree crowns within a 3D LiDAR scan of tropical forests, with only 2D noisy manual delineations of crowns on RGB images available as a source of weak supervision. We propose a refinement algorithm designed to enhance the performance of existing unsupervised learning methodologies for the ITC segmentation problem.

Subject Added Entry-Topical Term  

Statistics.

Subject Added Entry-Topical Term  

Computer science.

Subject Added Entry-Topical Term  

Information science.

Index Term-Uncontrolled  

Active learning

Index Term-Uncontrolled  

Imperfect supervision

Index Term-Uncontrolled  

Machine learning

Index Term-Uncontrolled  

Sufficient dimension reduction

Index Term-Uncontrolled  

Weakly supervised learning

Added Entry-Corporate Name  

Columbia University Statistics

Host Item Entry  

Dissertations Abstracts International. 85-12A.

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:658679