Graph-Based Semi-Supervised Learning

by Subramanya, Amarnag

Type Book

Rate 0.0 /5

Details

Files

Reviews

Download

Title Statements

Full title :Graph-Based Semi-Supervised Learning

Summary

While labeled data is expensive to prepare, ever increasing amounts of unlabeled data is becoming widely available. In order to adapt to this phenomenon, several semi-supervised learning (SSL) algorithms, which learn from labeled as well as unlabeled data, have been developed. In a separate line of work, researchers have started to realize that graphs provide a natural way to represent data in a variety of domains. Graph-based SSL algorithms, which bring together these two lines of work, have been shown to outperform the state-of-the-art in many applications in speech processing, computer vision, natural language processing, and other areas of Artificial Intelligence. Recognizing this promising and emerging area of research, this synthesis lecture focuses on graph based SSL algorithms (e.g., label propagation methods). Our hope is that after reading this book, the reader will walk away with the following: (1) an in-depth knowledge of the current state-of- the-art in graph-based SSL algorithms, and the ability to implement them; (2) the ability to decide on the suitability of graph-based SSL methods for a problem; and (3) familiarity with different applications where graph-based SSL methods have been successfully applied.

Publication Statements

Publisher(s):Morgan & Claypool Publishers
Language:eng

The Physical Object

Description:1 online resource

Edition Identifiers

ISBN:9781627052023, 162705202X, 9781627052016, 1627052011
oclc:on1229337205
LC Classification:Q
Dewey Decimal:006.31

1. Introduction -- 1.1 Unsupervised learning -- 1.2 Supervised learning -- 1.3 Semi-supervised learning (SSL) -- 1.4 Graph-based semi-supervised learning -- 1.4.1 Inductive vs. transductive SSL -- 1.5 Book organization

2. Graph construction -- 2.1 Problem statement -- 2.2 Task-independent graph construction -- 2.2.1 K-nearest neighbor (k-NN) and -neighborhood methods -- 2.2.2 Graph construction using b-matching -- 2.2.3 Graph construction using local reconstruction -- 2.3 Task-dependent graph construction -- 2.3.1 Inference-driven metric learning (IDML) -- 2.3.2 Graph kernels by spectral transform -- 2.4 Conclusion

3. Learning and inference -- 3.1 Seed supervision -- 3.2 Transductive methods -- 3.2.1 Graph cut -- 3.2.2 Gaussian random fields (GRF) -- 3.2.3 Local and global consistency (LGC) -- 3.2.4 Adsorption -- 3.2.5 Modified adsorption (MAD) -- 3.2.6 Quadratic criteria (QC) -- 3.2.7 Transduction with confidence (TACO) -- 3.2.8 Information regularization -- 3.2.9 Measure propagation -- 3.3 Inductive methods -- 3.3.1 Manifold regularization -- 3.4 Results on benchmark SSL data sets -- 3.5 Conclusions

4. Scalability -- 4.1 Large-scale graph construction -- 4.1.1 Approximate nea…