Partially labeled classification with Markov random walks

By Martin Szummer, Tommi S. Jaakkola. Neural Information Processing Systems, 2001

694 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/8e6779bb55f7fbed5684ded55df51747ea678a84

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Introduction to Semi-Supervised Learning (2009, methodology): “THE ASSUMPTION OF GRAPH-BASED METHODS∗ 55 walk [8, 168], harmonic function [210], local and global consistency [198], manifold regularization [17, 158, 155], kernels from the graph Laplacian [35, 51, 95, 101, 163, 211], spectral graph transducer [90], local averaging [179, 187], density-based regula”
Diffusion Kernels on Graphs and Other Discrete Input Spaces (2002, methodology): “In all of these cases, the challenge is to capture in the kernel the local and global structure of the graph.”
SemiBoost: Boosting for Semi-Supervised Learning (2009, methodology): “A continuous relaxation of this objective function, where yi ∈ [0, 1] has been considered in several approaches, which is solved using Markov random fields [5], Gaussian random fields and harmonic functions [18].”
Semi-supervised Protein Classification Using Cluster Kernels (2003, background): “Recent work in semi-supervised learning has focused on changing the representation given to a classifier by taking into account the structure described by the unlabeled data (Zhu and Ghahramani, 2002; Chapelle et al., 2002; Szummer and Jaakkola, 2001).”
A Framework for Learning Predictive Structures from Multiple Tasks and Unlabeled Data (2005, methodology): “An example of this approach is to use unlabeled data to create a data-manifold (graph structure), on which proper smooth function classes can be defined (Szummer and Jaakkola, 2002; Zhou et al., 2004; Zhu et al., 2003).”
Learning query intent from regularized click graphs (2008, methodology): “The problem of learning from labeled and unlabeled data has been investigated in a number of works [17, 21, 20, 12].”
Semi-supervised learning with graphs (2005, methodology): “Some Other Methods Szummer and Jaakkola (2001) perform a t-step Markov random walk on the graph.”
Video suggestion and discovery for youtube: taking random walks through the view graph (2008, background): “Based on the user–video view graph, if we recommend to user u a video v that she has not watched and that is closest in terms of graph distance, we may end up recommending a node that can only be reached via high-degree nodes (for example, if users u1 and u2 both watched a popular video, even…”
Semi-Supervised Learning for Natural Language (2005, methodology): “Other methods based on similarity between example pairs include label propagation (Zhu and Ghahramani, 2002), random walks (Szummer and Jaakkola, 2001), and spectral methods (Joachims, 2003).”
Semi-Supervised Classification of Network Data Using Very Few Labels (2010, background): “For some existing state-of-the-art semi-supervised learning methods the labeled data needed is reduced by a factor of 50.”
Reliable Data Distillation on Graph Convolutional Network (2020, methodology): “For example, we can learn a smooth low-dimensional embedding for each node with Markov random walks [45] and spectral kernels [58].”
Label Efficient Semi-Supervised Learning via Graph Filtering (2019, background): “One approach is to learn smooth low-dimensional embeddings with Markov random walks [45], Laplacian eigenmaps [3], spectral kernels [14, 57], and context-based methods [40].”
Diffusion maps and coarse-graining: a unified framework for dimensionality reduction, graph partitioning, and data set parameterization (2006, background): “In addition, the diffusion framework makes it possible to directly connect grouping in embedding spaces to spectral graph clustering and data analysis by Markov chains [21], [11].”
Semi-Supervised Learning on Riemannian Manifolds (2004, methodology): “While there has been recent work on using manifold structure for data representation (Roweis and Saul, 2000; Tenenbaum, et al, 2000), the only other application to classification, that we are aware of, was in (Szummer and Jaakkola, 2002), where the authors use a random walk on the adjacency graph fo”
Label Propagation via Teaching-to-Learn and Learning-to-Teach (2017, background): “The element P ij shows the probability of a particle jumping from node j to i in the random walks interpretation [5], [6], [39].”
New Regularized Algorithms for Transductive Learning (2009, background): “As many other related algorithms [17,12,5], Adsorption assumes that the learning problem is given in a graph form, where examples or instances are represented as nodes or vertices and edges code similarity between examples.”