Euclidean Embedding of Co-occurrence Data

By Amir Globerson, Gal Chechik, Fernando C Pereira, Naftali Tishby. 2004

267 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Representation & Self-Supervised Learning, Signal Processing

How researchers see this paper

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

In Artificial Intelligence

In Information Systems

In Representation & Self-Supervised Learning

In Signal Processing

Citations in the corpus

Rethinking LDA: Moment Matching for Discrete ICA (2015, neutral): “Blei’s web page,4 withN = 2, 243 documents and M = 10, 473 vocabulary words and the average document length L̂ = 194; the NIPS papers dataset5 [28] of 2, 483 NIPS papers and 14, 036 words, and L̂ = 1, 321; the KOS dataset,6 from the UCI Repository, with 3, 430 documents and 6, 906 words, and L̂ = 13”
SourceFinder: Finding Malware Source-Code from Publicly Available Repositories (2020, background): “Furthermore, the word co-occurrences does not exist in topics and ﬁle names ﬁeld, which partly what makes embedding approaches work well in large and well structured documents [25,40].”
DeepTingle (2017, methodology): “Such mappings have been achieved using various techniques, such as neural networks (Bengio, Ducharme, and Vincent 2003), principal component analysis (Lebret and Collobert 2013), and probabilistic models (Globerson et al. 2007).”
Learning a Parametric Embedding by Preserving Local Structure (2009, methodology): “It may seem logical to use a Gaussian distribution to measure pairwise similarities q ij in the latent space (as is done in, e.g., (Globerson et al., 2007, Hinton, 2002, Min, 2005, Iwata et al., 2007)), but this often leads to inferior results that are due to the crowding problem (van der Maaten…”
Learning Syntactic Categories Using Paradigmatic Representations of Word Context (2012, methodology): “In this section we review the unsupervised meth-ods we used to model co-occurrence statistics: the Co-occurrence Data Embedding (CODE) method (Globerson et al., 2007) and its spherical extension (S-CODE) introduced by (Maron et al., 2010).”
Input Output Kernel Regression (2016, methodology): “We applied our method on a co-authorship network containing information on publications of the NIPS conferences between 1988 to 2003 (Globerson et al., 2007).”
Deep Learning for Entity Matching: A Design Space Exploration (2018, methodology): “Methods to generate these mappings include neural networks [52], dimensionality reduction techniques such as PCA [45] on the word co-occurrence matrix and other probabilistic techniques [25].”
On Deep Multi-View Representation Learning (2015, methodology): “For multi-view data, Globerson et al. (2007) propose an algorithm for learning Euclidean embeddings by deﬁning a joint or conditional distribution of the views based on Euclidean distance in the embedding space and maximizing the data likelihood.”
Large-Scale Embedding Learning in Heterogeneous Event Data (2016, background): “The objective of embedding techniques is mainly to preserve certain relationships among objects [5]–[12].”
RAND-WALK: A Latent Variable Model Approach to Word Embeddings (2015, background): “(2) Variants of the expression for joint probability in (2.3) had been hypothesized based upon empirics in (Mikolov et al., 2013b) and also (Globerson et al., 2007), and (Maron et al., 2010) .”
A Latent Variable Model Approach to PMI-based Word Embeddings (2015, background): “Variants of the expression for joint probability in (2.3) had been hypothesized based upon empirical evidence in Mikolov et al. (2013b) and also Globerson et al. (2007), and Maron et al. (2010) .”
Dimensionality Reduction of Multimodal Labeled Data by Local Fisher Discriminant Analysis (2007, background): “There are several measures for quantitatively evaluating separability of data samples in different classes (e.g., Fukunaga, 1990; Globerson et al., 2005).”
Semi-supervised Penalized Output Kernel Regression for Link Prediction (2011, unknown): “We apply our method on a co-authorship dataset (Globerson et al., 2007) containing information on publications of the NIPS conferences from 1988 to 2003.”