arXiv subjects

Waleed Ammar

Publications and source records attributed to Waleed Ammar.

22 records · Page 2Linked to original sources

arXiv2016

Massively Multilingual Word Embeddings

We introduce new methods for estimating and evaluating embeddings of words in more than fifty languages in a single shared embedding space. Our estimation methods, multiCluster and multiCCA, use dictionaries and monolingual data; they do not require parallel data. Our new evaluation method, multiQVEC-CCA, is shown to correlate better than previous ones with two downstream tasks (text categorization and parsing). We also describe a web portal for evaluation that will facilitate further research in this area, along with open-source releases of all our methods.

cs.CL↗

arXiv2015

Unsupervised POS Induction with Word Embeddings

Unsupervised word embeddings have been shown to be valuable as features in supervised learning problems; however, their role in unsupervised problems has been less thoroughly explored. In this paper, we show that embeddings can likewise add value to the problem of unsupervised POS induction. In two representative models of POS induction, we replace multinomial distributions over the vocabulary with multivariate Gaussian distributions over word embeddings and observe consistent improvements in eight languages. We also analyze the effect of various choices while inducing word embeddings on "downstream" POS induction results.

cs.CL↗

arXiv2014

Conditional Random Field Autoencoders for Unsupervised Structured Prediction

We introduce a framework for unsupervised learning of structured predictors with overlapping, global features. Each input's latent representation is predicted conditional on the observable data using a feature-rich conditional random field. Then a reconstruction of the input is (re)generated, conditional on the latent structure, using models for which maximum likelihood estimation has a closed-form. Our autoencoder formulation enables efficient learning without making unrealistic independence assumptions or restricting the kinds of features that can be used. We illustrate insightful connections to traditional autoencoders, posterior regularization and multi-view learning. We show competitive results with instantiations of the model for two canonical NLP tasks: part-of-speech induction and bitext word alignment, and show that training our model can be substantially more efficient than comparable feature-rich baselines.

cs.LG↗

arXiv2010

Secure Localization in Wireless Sensor Networks: A Survey

Wireless sensor networks (WSNs) have gained researchers' attention in the last several years. Small sensors powered by miniaturized microprocessors are capable of supporting several applications for civil and military domains. Determining the location of sensors is a basic and essential knowledge for most WSN algorithms and protocols including data tagging, routing, node identification, among others. This paper surveys the different algorithms that have been proposed to securely determine the location of a sensor node. By "secure", we mean that adversaries cannot easily affect the accuracy of the localized sensors. In other words, the localization algorithm must be robust under several attacks. We provide a taxonomy for classifying different secure localization schemes and describe possible attacks that can harm localization. In addition, we survey different secure localization schemes and show how they map to the proposed taxonomy. We also give a comparison between the different schemes, showing the attacks addressed by each.

cs.CR↗