Filters








56 Hits in 1.3 sec

Revealing the Basis: Ordinal Embedding Through Geometry [article]

Jesse Anderton, Virgil Pavlu, Javed Aslam
2018 arXiv   pre-print
Ordinal Embedding places n objects into R^d based on comparisons such as "a is closer to b than c." Current optimization-based approaches suffer from scalability problems and an abundance of low quality local optima. We instead consider a computational geometric approach based on selecting comparisons to discover points close to nearly-orthogonal "axes" and embed the whole set by their projections along each axis. We thus also estimate the dimensionality of the data. Our embeddings are of lower
more » ... quality than the global optima of optimization-based approaches, but are more scalable computationally and more reliable than local optima often found via optimization. Our method uses \Theta(n d \log n) comparisons and \Theta(n^2 d^2) total operations, and can also be viewed as selecting constraints for an optimizer which, if successful, will produce an almost-perfect embedding for sufficiently dense datasets.
arXiv:1805.07589v1 fatcat:fcrorh3safhgddcpodvm2o3et4

Measuring Human-perceived Similarity in Heterogeneous Collections [article]

Jesse Anderton, Pavel Metrikov, Virgil Pavlu, Javed Aslam
2018 arXiv   pre-print
We present a technique for estimating the similarity between objects such as movies or foods whose proper representation depends on human perception. Our technique combines a modest number of human similarity assessments to infer a pairwise similarity function between the objects. This similarity function captures some human notion of similarity which may be difficult or impossible to automatically extract, such as which movie from a collection would be a better substitute when the desired one
more » ... s unavailable. In contrast to prior techniques, our method does not assume that all similarity questions on the collection can be answered or that all users perceive similarity in the same way. When combined with a user model, we find how each assessor's tastes vary, affecting their perception of similarity.
arXiv:1802.05929v1 fatcat:iauektghivcrrcz6nzxqtchwka

TREC 2013 Temporal Summarization

Javed A. Aslam, Matthew Ekstrand-Abueg, Virgil Pavlu, Fernando Diaz, Tetsuya Sakai
2013 Text Retrieval Conference  
dblp:conf/trec/AslamEPDS13 fatcat:tsu6p7xjcvhwbezznnukappw3u

Adapting RNN Sequence Prediction Model to Multi-label Set Prediction [article]

Kechen Qin, Cheng Li, Virgil Pavlu, Javed A. Aslam
2019 arXiv   pre-print
We present an adaptation of RNN sequence models to the problem of multi-label classification for text, where the target is a set of labels, not a sequence. Previous such RNN models define probabilities for sequences but not for sets; attempts to obtain a set probability are after-thoughts of the network design, including pre-specifying the label order, or relating the sequence probability to the set probability in ad hoc ways. Our formulation is derived from a principled notion of set
more » ... y, as the sum of probabilities of corresponding permutation sequences for the set. We provide a new training objective that maximizes this set probability, and a new prediction objective that finds the most probable set on a test document. These new objectives are theoretically appealing because they give the RNN model freedom to discover the best label order, which often is the natural one (but different among documents). We develop efficient procedures to tackle the computation difficulties involved in training and prediction. Experiments on benchmark datasets demonstrate that we outperform state-of-the-art methods for this task.
arXiv:1904.05829v1 fatcat:7ga5knttf5anbfg4ze4gm3dog4

Adapting

Kechen Qin, Cheng Li, Virgil Pavlu, Javed Aslam
2019 Proceedings of the 2019 Conference of the North  
We present an adaptation of RNN sequence models to the problem of multi-label classification for text, where the target is a set of labels, not a sequence. Previous such RNN models define probabilities for sequences but not for sets; attempts to obtain a set probability are after-thoughts of the network design, including pre-specifying the label order, or relating the sequence probability to the set probability in ad hoc ways. Our formulation is derived from a principled notion of set
more » ... y, as the sum of probabilities of corresponding permutation sequences for the set. We provide a new training objective that maximizes this set probability, and a new prediction objective that finds the most probable set on a test document. These new objectives are theoretically appealing because they give the RNN model freedom to discover the best label order, which often is the natural one (but different among documents). We develop efficient procedures to tackle the computation difficulties involved in training and prediction. Experiments on benchmark datasets demonstrate that we outperform stateof-the-art methods for this task.
doi:10.18653/v1/n19-1321 dblp:conf/naacl/QinLPA19 fatcat:vnea377afbdy3pquzmuqcs4nsi

Regularizing Model Complexity and Label Structure for Multi-Label Text Classification [article]

Bingyu Wang, Cheng Li, Virgil Pavlu, Javed Aslam
2017 arXiv   pre-print
Multi-label text classification is a popular machine learning task where each document is assigned with multiple relevant labels. This task is challenging due to high dimensional features and correlated labels. Multi-label text classifiers need to be carefully regularized to prevent the severe over-fitting in the high dimensional space, and also need to take into account label dependencies in order to make accurate predictions under uncertainty. We demonstrate significant and practical
more » ... nt by carefully regularizing the model complexity during training phase, and also regularizing the label search space during prediction phase. Specifically, we regularize the classifier training using Elastic-net (L1+L2) penalty for reducing model complexity/size, and employ early stopping to prevent overfitting. At prediction time, we apply support inference to restrict the search space to label sets encountered in the training set, and F-optimizer GFM to make optimal predictions for the F1 metric. We show that although support inference only provides density estimations on existing label combinations, when combined with GFM predictor, the algorithm can output unseen label combinations. Taken collectively, our experiments show state of the art results on many benchmark datasets. Beyond performance and practical contributions, we make some interesting observations. Contrary to the prior belief, which deems support inference as purely an approximate inference procedure, we show that support inference acts as a strong regularizer on the label prediction structure. It allows the classifier to take into account label dependencies during prediction even if the classifiers had not modeled any label dependencies during training.
arXiv:1705.00740v1 fatcat:374ssbwjmvb25duau7wkwko5dy

Variational bayes for modeling score distributions

Keshi Dai, Evangelos Kanoulas, Virgil Pavlu, Javed A. Aslam
2010 Information retrieval (Boston)  
Empirical modeling of the score distributions associated with retrieved documents is an essential task for many retrieval applications. In this work, we propose modeling the relevant documents' scores by a mixture of Gaussians and the non-relevant scores by a Gamma distribution. Applying Variational Bayes we automatically trade-off the goodness-of-fit with the complexity of the model. We test our model on traditional retrieval functions and actual search engines submitted to TREC. We
more » ... the utility of our model in inferring precision-recall curves. In all experiments our model outperforms the dominant exponential-Gaussian model.
doi:10.1007/s10791-010-9156-2 fatcat:3sg2i3ggbbbkrpyjiigk2joagi

A Complex KBQA System using Multiple Reasoning Paths [article]

Kechen Qin, Yu Wang, Cheng Li, Kalpa Gunaratna, Hongxia Jin, Virgil Pavlu, Javed A. Aslam
2020 arXiv   pre-print
Multi-hop knowledge based question answering (KBQA) is a complex task for natural language understanding. Many KBQA approaches have been proposed in recent years, and most of them are trained based on labeled reasoning path. This hinders the system's performance as many correct reasoning paths are not labeled as ground truth, and thus they cannot be learned. In this paper, we introduce an end-to-end KBQA system which can leverage multiple reasoning paths' information and only requires labeled
more » ... swer as supervision. We conduct experiments on several benchmark datasets containing both single-hop simple questions as well as muti-hop complex questions, including WebQuestionSP (WQSP), ComplexWebQuestion-1.1 (CWQ), and PathQuestion-Large (PQL), and demonstrate strong performance.
arXiv:2005.10970v1 fatcat:zcn4yk4w4jgy7oqu3vnfz2oaha

Northeastern University Runs at the TREC13 Crowdsourcing Track

Maryam Bashir, Jesse Anderton, Virgil Pavlu, Javed A. Aslam
2013 Text Retrieval Conference  
The goal of the TREC 2013 Crowdsourcing Track was to evaluate approaches to crowdsourcing high quality relevance judgments for web pages and search topics. This paper describes our submission to Crowdsourcing track. Participants of this track were required to assess documents judged on a six-point scale. Our approach is based on collecting a linear number of preference judgements, and combining these into nominal grades using a modified version of QuickSort algorithm.
dblp:conf/trec/BashirAPA13 fatcat:ejfkuk4gqbaehisqaa7cpmr4ki

If I Had a Million Queries [chapter]

Ben Carterette, Virgil Pavlu, Evangelos Kanoulas, Javed A. Aslam, James Allan
2009 Lecture Notes in Computer Science  
As document collections grow larger, the information needs and relevance judgments in a test collection must be well-chosen within a limited budget to give the most reliable and robust evaluation results. In this work we analyze a sample of queries categorized by length and corpus-appropriateness to determine the right proportion needed to distinguish between systems. We also analyze the appropriate division of labor between developing topics and making relevance judgments, and show that only a
more » ... small, biased sample of queries with sparse judgments is needed to produce the same results as a much larger sample of queries.
doi:10.1007/978-3-642-00958-7_27 fatcat:ikc7siy45jcj5a5y32fnnoiy5u

TREC 2015 Temporal Summarization Track Overview

Javed A. Aslam, Fernando Diaz, Matthew Ekstrand-Abueg, Richard McCreadie, Virgil Pavlu, Tetsuya Sakai
2015 Text Retrieval Conference  
dblp:conf/trec/AslamDEMPS15 fatcat:ljksoyzqwvfwhi6eeuv4tjh77q

TREC 2014 Temporal Summarization Track Overview

Javed A. Aslam, Matthew Ekstrand-Abueg, Virgil Pavlu, Fernando Diaz, Richard McCreadie, Tetsuya Sakai
2014 Text Retrieval Conference  
dblp:conf/trec/AslamEPDMS14 fatcat:a3ql3eaxwvauxp36ovclx4vqbu

Extended Expectation Maximization for Inferring Score Distributions [chapter]

Keshi Dai, Virgil Pavlu, Evangelos Kanoulas, Javed A. Aslam
2012 Lecture Notes in Computer Science  
Inferring the distributions of relevant and nonrelevant documents over a ranked list of scored documents returned by a retrieval system has a broad range of applications including information filtering, recall-oriented retrieval, metasearch, and distributed IR. Typically, the distribution of documents over scores is modeled by a mixture of two distributions, one for the relevant and one for the nonrelevant documents, and expectation maximization (EM) is run to estimate the mixture parameters. A
more » ... large volume of work has focused on selecting the appropriate form of the two distributions in the mixture. In this work we consider the form of the distributions as a given and we focus on the inference algorithm. We extend the EM algorithm (a) by simultaneously considering the ranked lists of documents returned by multiple retrieval systems, and (b) by encoding in the algorithm the constraint that the same document retrieved by multiple systems should have the same, global, probability of relevance. We test the new inference algorithm using TREC data and we demonstrate that it outperforms the regular EM algorithm. It is better calibrated in inferring the probability of document's relevance, and it is more effective when applied on the task of metasearch.
doi:10.1007/978-3-642-28997-2_25 fatcat:arugdpwpc5grnk6iwhzf3jzvuu

Conditional Bernoulli Mixtures for Multi-label Classification

Cheng Li, Bingyu Wang, Virgil Pavlu, Javed A. Aslam
2016 International Conference on Machine Learning  
Multi-label classification is an important machine learning task wherein one assigns a subset of candidate labels to an object. In this paper, we propose a new multi-label classification method based on Conditional Bernoulli Mixtures. Our proposed method has several attractive properties: it captures label dependencies; it reduces the multi-label problem to several standard binary and multi-class problems; it subsumes the classic independent binary prediction and power-set subset prediction
more » ... ods as special cases; and it exhibits accuracy and/or computational complexity advantages over existing approaches. We demonstrate two implementations of our method using logistic regressions and gradient boosted trees, together with a simple training procedure based on Expectation Maximization. We further derive an efficient prediction procedure based on dynamic programming, thus avoiding the cost of examining an exponential number of potential label subsets. Experimental results show the effectiveness of the proposed method against competitive alternatives on benchmark datasets.
dblp:conf/icml/LiWPA16 fatcat:lrmr56w3z5eudfqyfvcnj5yo6q

Northeastern University Runs at the TREC12 Crowdsourcing Track

Maryam Bashir, Jesse Anderton, Jie Wu, Matthew Ekstrand-Abueg, Peter B. Golbus, Virgil Pavlu, Javed A. Aslam
2012 Text Retrieval Conference  
The goal of the TREC 2012 Crowdsourcing Track was to evaluate approaches to crowdsourcing high quality relevance judgments for images and text documents. This paper describes our submission to the Text Relevance Assessing Task. We explored three different approaches for obtaining relevance judgments. Our first two approaches are based on collecting a limited number of preference judgments from Amazon Mechanical Turk workers. These preferences are then extended to relevance judgments through the
more » ... use of expectation maximization and the Elo rating system. Our third approach is based on our Nugget-based evaluation paradigm.
dblp:conf/trec/BashirAWEGPA12 fatcat:v2ja3fo7wfgwxamffgyjc7nrsm
« Previous Showing results 1 — 15 out of 56 results