arXiv subjects

Pawan Goyal

Publications and source records attributed to Pawan Goyal.

At least 163 records · Page 9Linked to original sources

arXiv2020

Low-Rank and Total Variation Regularization and Its Application to Image Recovery

In this paper, we study the problem of image recovery from given partial (corrupted) observations. Recovering an image using a low-rank model has been an active research area in data analysis and machine learning. But often, images are not only of low-rank but they also exhibit sparsity in a transformed space. In this work, we propose a new problem formulation in such a way that we seek to recover an image that is of low-rank and has sparsity in a transformed domain. We further discuss various non-convex non-smooth surrogates of the rank function, leading to a relaxed problem. Then, we present an efficient iterative scheme to solve the relaxed problem that essentially employs the (weighted) singular value thresholding at each iteration. Furthermore, we discuss the convergence properties of the proposed iterative method. We perform extensive experiments, showing that the proposed algorithm outperforms state-of-the-art methodologies in recovering images.

cs.CV↗

arXiv2020

Using Distributional Thesaurus Embedding for Co-hyponymy Detection

Discriminating lexical relations among distributionally similar words has always been a challenge for natural language processing (NLP) community. In this paper, we investigate whether the network embedding of distributional thesaurus can be effectively utilized to detect co-hyponymy relations. By extensive experiments over three benchmark datasets, we show that the vector representation obtained by applying node2vec on distributional thesaurus outperforms the state-of-the-art models for binary classification of co-hyponymy vs. hypernymy, as well as co-hyponymy vs. meronymy, by huge margins.

cs.CL↗

arXiv2019

Analysis of Reference and Citation Copying in Evolving Bibliographic Networks

Extensive literature demonstrates how the copying of references (links) can lead to the emergence of various structural properties (e.g., power-law degree distribution and bipartite cores) in bibliographic and other similar directed networks. However, it is also well known that the copying process is incapable of mimicking the number of directed triangles in such networks; neither does it have the power to explain the obsolescence of older papers. In this paper, we propose RefOrCite, a new model that allows for copying of both the references from (i.e., out-neighbors of) as well as the citations to (i.e., in-neighbors of) an existing node. In contrast, the standard copying model (CP) only copies references. While retaining its spirit, RefOrCite differs from the Forest Fire (FF) model in ways that makes RefOrCite amenable to mean-field analysis for degree distribution, triangle count, and densification. Empirically, RefOrCite gives the best overall agreement with observed degree distribution, triangle count, diameter, h-index, and the growth of citations to newer papers.

cs.SI↗

arXiv2019

Weakly-Supervised Deep Learning for Domain Invariant Sentiment Classification

The task of learning a sentiment classification model that adapts well to any target domain, different from the source domain, is a challenging problem. Majority of the existing approaches focus on learning a common representation by leveraging both source and target data during training. In this paper, we introduce a two-stage training procedure that leverages weakly supervised datasets for developing simple lift-and-shift-based predictive models without being exposed to the target domain during the training phase. Experimental results show that transfer with weak supervision from a source domain to various target domains provides performance very close to that obtained via supervised training on the target domain itself.

cs.LG↗

arXiv2019

Identification of Port-Hamiltonian Systems from Frequency Response Data

In this paper, we study the identification problem of a passive system from tangential interpolation data. We present a simple construction approach based on the Mayo-Antoulas generalized realization theory that automatically yields a port-Hamiltonian realization for every strictly passive system with simple spectral zeros. Furthermore, we discuss the construction of a frequency-limited port-Hamiltonian realization. We illustrate the proposed method by means of several examples.

eess.SY↗

arXiv2019

Identification of Dominant Subspaces for Linear Structured Parametric Systems and Model Reduction

In this paper, we discuss a novel model reduction framework for generalized linear systems. The transfer functions of these systems are assumed to have a special structure, e.g., coming from second-order linear systems and time-delay systems, and they may also have parameter dependencies. Firstly, we investigate the connection between classic interpolation-based model reduction methods with the reachability and observability subspaces of linear structured parametric systems. We show that if enough interpolation points are taken, the projection matrices of interpolation-based model reduction encode these subspaces. As a result, we are able to identify the dominant reachable and observable subspaces of the underlying system. Based on this, we propose a new model reduction algorithm combining these features leading to reduced-order systems. Furthermore, we pay special attention to computational aspects of the approach and discuss its applicability to a large-scale setting. We illustrate the efficiency of the proposed approach with several numerical large-scale benchmark examples.

math.NA↗

arXiv2019

Data-Driven Identification of Rayleigh-Damped Second-Order Systems

In this paper, we present a data-driven approach to identify second-order systems, having internal Rayleigh damping. This means that the damping matrix is given as a linear combination of the mass and stiffness matrices. These systems typically appear when performing various engineering studies, e.g., vibrational and structural analysis. In an experimental setup, the frequency response of a system can be measured via various approaches, for instance, by measuring the vibrations using an accelerometer. As a consequence, given frequency samples, the identification of the underlying system relies on rational approximation. To that aim, we propose an identification of the corresponding second-order system, extending the Loewner framework for this class of systems. The efficiency of the proposed method is demonstrated by means of various numerical benchmarks.

math.OC↗

arXiv2019

HateMonitors: Language Agnostic Abuse Detection in Social Media

Reducing hateful and offensive content in online social media pose a dual problem for the moderators. On the one hand, rigid censorship on social media cannot be imposed. On the other, the free flow of such content cannot be allowed. Hence, we require efficient abusive language detection system to detect such harmful content in social media. In this paper, we present our machine learning model, HateMonitor, developed for Hate Speech and Offensive Content Identification in Indo-European Languages (HASOC), a shared task at FIRE 2019. We have used a Gradient Boosting model, along with BERT and LASER embeddings, to make the system language agnostic. Our model came at First position for the German sub-task A. We have also made our model public at https://github.com/punyajoy/HateMonitors-HASOC .

cs.SI↗

arXiv2019

Competing Topic Naming Conventions in Quora: Predicting Appropriate Topic Merges and Winning Topics from Millions of Topic Pairs

Quora is a popular Q&A site which provides users with the ability to tag questions with multiple relevant topics which helps to attract quality answers. These topics are not predefined but user-defined conventions and it is not so rare to have multiple such conventions present in the Quora ecosystem describing exactly the same concept. In almost all such cases, users (or Quora moderators) manually merge the topic pair into one of the either topics, thus selecting one of the competing conventions. An important application for the site therefore is to identify such competing conventions early enough that should merge in future. In this paper, we propose a two-step approach that uniquely combines the anomaly detection and the supervised classification frameworks to predict whether two topics from among millions of topic pairs are indeed competing conventions, and should merge, achieving an F-score of 0.711. We also develop a model to predict the direction of the topic merge, i.e., the winning convention, achieving an F-score of 0.898. Our system is also able to predict ~ 25% of the correct case of merges within the first month of the merge and ~ 40% of the cases within a year. This is an encouraging result since Quora users on average take 936 days to identify such a correct merge. Human judgment experiments show that our system is able to predict almost all the correct cases that humans can predict plus 37.24% correct cases which the humans are not able to identify at all.

cs.SI↗

arXiv2019

Gramians, Energy Functionals and Balanced Truncation for Linear Dynamical Systems with Quadratic Outputs

Model order reduction is a technique that is used to construct low-order approximations of large-scale dynamical systems. In this paper, we investigate a balancing based model order reduction method for dynamical systems with a linear dynamical equation and a quadratic output function. To this aim, we propose a new algebraic observability Gramian for the system based on Hilbert space adjoint theory. We then show the proposed Gramians satisfy a particular type of generalized Lyapunov equations and we investigate their connections to energy functionals, namely, the controllability and observability. This allows us to find the states that are hard to control and hard to observe via an appropriate balancing transformation. Truncation of such states yields reduced-order systems. Finally, based on $\mathcal H_2$ energy considerations, we, furthermore, derive error bounds, depending on the neglected singular values. The efficiency of the proposed method is demonstrated by means of two semi-discretized partial differential equations and is compared with the existing model reduction techniques in the literature.

math.OC↗

arXiv2019

Incorporating Domain Knowledge into Medical NLI using Knowledge Graphs

Recently, biomedical version of embeddings obtained from language models such as BioELMo have shown state-of-the-art results for the textual inference task in the medical domain. In this paper, we explore how to incorporate structured domain knowledge, available in the form of a knowledge graph (UMLS), for the Medical NLI task. Specifically, we experiment with fusing embeddings obtained from knowledge graph with the state-of-the-art approaches for NLI task (ESIM model). We also experiment with fusing the domain-specific sentiment information for the task. Experiments conducted on MedNLI dataset clearly show that this strategy improves the baseline BioELMo architecture for the Medical NLI task.

cs.CL↗

arXiv2019

The rise and rise of interdisciplinary research: Understanding the interaction dynamics of three major fields -- Physics, Mathematics & Computer Science

The distinction between sciences is becoming increasingly more artificial -- an approach from one area can be easily applied to the other. More exciting research nowadays is happening perhaps at the interfaces of disciplines like Physics, Mathematics and Computer Science. How do these interfaces emerge and interact? For instance, is there a specific pattern in which these fields cite each other? In this article, we investigate a collection of more than 1.2 million papers from three different scientific disciplines -- Physics, Mathematics, and Computer Science. We show how over a timescale the citation patterns from the core science fields (Physics, Mathematics) to the applied and fast-growing field of Computer Science have drastically increased. Further, we observe how certain subfields in these disciplines are shrinking while others are becoming tremendously popular. For instance, an intriguing observation is that citations from Mathematics to the subfield of machine learning in Computer Science in recent times are exponentially increasing.

cs.DL↗

arXiv2019

On the Compositionality Prediction of Noun Phrases using Poincaré Embeddings

The compositionality degree of multiword expressions indicates to what extent the meaning of a phrase can be derived from the meaning of its constituents and their grammatical relations. Prediction of (non)-compositionality is a task that has been frequently addressed with distributional semantic models. We introduce a novel technique to blend hierarchical information with distributional information for predicting compositionality. In particular, we use hypernymy information of the multiword and its constituents encoded in the form of the recently introduced Poincaré embeddings in addition to the distributional information to detect compositionality for noun phrases. Using a weighted average of the distributional similarity and a Poincaré similarity function, we obtain consistent and substantial, statistically significant improvement across three gold standard datasets over state-of-the-art models based on distributional information only. Unlike traditional approaches that solely use an unsupervised setting, we have also framed the problem as a supervised task, obtaining comparable improvements. Further, we publicly release our Poincaré embeddings, which are trained on the output of handcrafted lexical-syntactic patterns on a large corpus.

cs.CL↗

arXiv2019

Interpolation-Based Model Order Reduction for Polynomial Parametric Systems

In this work, we investigate a model order reduction scheme for polynomial parametric systems. We begin with defining the generalized multivariate transfer functions for the system. Based on this, we aim at constructing a reduced-order system, interpolating the defined generalized transfer functions at a given set of interpolation points. Furthermore, we provide a method, inspired by the Loewner approach for linear and (quadratic-)bilinear systems, to determine a good-quality reduced-order system in an automatic way. We also discuss the computational issues related to the proposed method and a potential application of CUR matrix approximation in order to further speed-up simulations of reduced-order systems. We test the efficiency of the proposed methods via several numerical examples.

math.NA↗

arXiv2019

Thou shalt not hate: Countering Online Hate Speech

Hate content in social media is ever-increasing. While Facebook, Twitter, Google have attempted to take several steps to tackle the hateful content, they have mostly been unsuccessful. Counterspeech is seen as an effective way of tackling the online hate without any harm to the freedom of speech. Thus, an alternative strategy for these platforms could be to promote counterspeech as a defense against hate content. However, in order to have a successful promotion of such counterspeech, one has to have a deep understanding of its dynamics in the online world. Lack of carefully curated data largely inhibits such understanding. In this paper, we create and release the first ever dataset for counterspeech using comments from YouTube. The data contains 13,924 manually annotated comments where the labels indicate whether a comment is a counterspeech or not. This data allows us to perform a rigorous measurement study characterizing the linguistic structure of counterspeech for the first time. This analysis results in various interesting insights such as: the counterspeech comments receive much more likes as compared to the non-counterspeech comments, for certain communities majority of the non-counterspeech comments tend to be hate speech, the different types of counterspeech are not all equally effective and the language choice of users posting counterspeech is largely different from those posting non-counterspeech as revealed by a detailed psycholinguistic analysis. Finally, we build a set of machine learning models that are able to automatically detect counterspeech in YouTube videos with an F1-score of 0.71. We also build multilabel models that can detect different types of counterspeech in a comment with an F1-score of 0.60.

cs.SI↗

arXiv2019

DeepTagRec: A Content-cum-User based Tag Recommendation Framework for Stack Overflow

In this paper, we develop a content-cum-user based deep learning framework DeepTagRec to recommend appropriate question tags on Stack Overflow. The proposed system learns the content representation from question title and body. Subsequently, the learnt representation from heterogeneous relationship between user and tags is fused with the content representation for the final tag prediction. On a very large-scale dataset comprising half a million question posts, DeepTagRec beats all the baselines; in particular, it significantly outperforms the best performing baseline T agCombine achieving an overall gain of 60.8% and 36.8% in precision@3 and recall@10 respectively. DeepTagRec also achieves 63% and 33.14% maximum improvement in exact-k accuracy and top-k accuracy respectively over TagCombine

cs.SI↗

arXiv2019

Automated Early Leaderboard Generation From Comparative Tables

A leaderboard is a tabular presentation of performance scores of the best competing techniques that address a specific scientific problem. Manually maintained leaderboards take time to emerge, which induces a latency in performance discovery and meaningful comparison. This can delay dissemination of best practices to non-experts and practitioners. Regarding papers as proxies for techniques, we present a new system to automatically discover and maintain leaderboards in the form of partial orders between papers, based on performance reported therein. In principle, a leaderboard depends on the task, data set, other experimental settings, and the choice of performance metrics. Often there are also tradeoffs between different metrics. Thus, leaderboard discovery is not just a matter of accurately extracting performance numbers and comparing them. In fact, the levels of noise and uncertainty around performance comparisons are so large that reliable traditional extraction is infeasible. We mitigate these challenges by using relatively cleaner, structured parts of the papers, e.g., performance tables. We propose a novel performance improvement graph with papers as nodes, where edges encode noisy performance comparison information extracted from tables. Every individual performance edge is extracted from a table with citations to other papers. These extractions resemble (noisy) outcomes of 'matches' in an incomplete tournament. We propose several approaches to rank papers from these noisy 'match' outcomes. We show that our ranking scheme can reproduce various manually curated leaderboards very well. Using widely-used lists of state-of-the-art papers in 27 areas of Computer Science, we demonstrate that our system produces very reliable rankings.

cs.DL↗

arXiv2019

Misleading Metadata Detection on YouTube

YouTube is the leading social media platform for sharing videos. As a result, it is plagued with misleading content that includes staged videos presented as real footages from an incident, videos with misrepresented context and videos where audio/video content is morphed. We tackle the problem of detecting such misleading videos as a supervised classification task. We develop UCNet - a deep network to detect fake videos and perform our experiments on two datasets - VAVD created by us and publicly available FVC [8]. We achieve a macro averaged F-score of 0.82 while training and testing on a 70:30 split of FVC, while the baseline model scores 0.36. We find that the proposed model generalizes well when trained on one dataset and tested on the other.

cs.CL↗