arXiv subjects

Ritesh Kumar

Publications and source records attributed to Ritesh Kumar.

At least 55 records · Page 3Linked to original sources

arXiv2021

SIGTYP 2021 Shared Task: Robust Spoken Language Identification

While language identification is a fundamental speech and language processing task, for many languages and language families it remains a challenging task. For many low-resource and endangered languages this is in part due to resource availability: where larger datasets exist, they may be single-speaker or have different domains than desired application scenarios, demanding a need for domain and speaker-invariant language identification systems. This year's shared task on robust spoken language identification sought to investigate just this scenario: systems were to be trained on largely single-speaker speech from one domain, but evaluated on data in other domains recorded from speakers under different recording circumstances, mimicking realistic low-resource scenarios. We see that domain and speaker mismatch proves very challenging for current methods which can perform above 95% accuracy in-domain, which domain adaptation can address to some degree, but that these conditions merit further investigation to make spoken language identification accessible in many scenarios.

cs.CL↗

arXiv2021

Unsupervised Learning of Explainable Parse Trees for Improved Generalisation

Recursive neural networks (RvNN) have been shown useful for learning sentence representations and helped achieve competitive performance on several natural language inference tasks. However, recent RvNN-based models fail to learn simple grammar and meaningful semantics in their intermediate tree representation. In this work, we propose an attention mechanism over Tree-LSTMs to learn more meaningful and explainable parse tree structures. We also demonstrate the superior performance of our proposed model on natural language inference, semantic relatedness, and sentiment analysis tasks and compare them with other state-of-the-art RvNN based methods. Further, we present a detailed qualitative and quantitative analysis of the learned parse trees and show that the discovered linguistic structures are more explainable, semantically meaningful, and grammatically correct than recent approaches. The source code of the paper is available at https://github.com/atul04/Explainable-Latent-Structures-Using-Attention.

cs.CL↗

arXiv2021

MSTREAM: Fast Anomaly Detection in Multi-Aspect Streams

Given a stream of entries in a multi-aspect data setting i.e., entries having multiple dimensions, how can we detect anomalous activities in an unsupervised manner? For example, in the intrusion detection setting, existing work seeks to detect anomalous events or edges in dynamic graph streams, but this does not allow us to take into account additional attributes of each entry. Our work aims to define a streaming multi-aspect data anomaly detection framework, termed MSTREAM which can detect unusual group anomalies as they occur, in a dynamic manner. MSTREAM has the following properties: (a) it detects anomalies in multi-aspect data including both categorical and numeric attributes; (b) it is online, thus processing each record in constant time and constant memory; (c) it can capture the correlation between multiple aspects of the data. MSTREAM is evaluated over the KDDCUP99, CICIDS-DoS, UNSW-NB 15 and CICIDS-DDoS datasets, and outperforms state-of-the-art baselines.

cs.LG↗

arXiv2020

Non-ballistic transport characteristics of superconducting point-contacts

In the "ballistic" regime, the transport across a normal metal (N)/superconductor (S) point-contact is dominated by a quantum process called Andreev reflection. Andreev reflection causes an enhancement of the conductance below the superconducting energy gap, and the ratio of the low-bias and the high-bias conductance cannot be greater than 2 when the superconductor is conventional in nature. In this regime, the features associated with Andreev reflection also provide energy and momentum-resolved spectroscopic information about the superconducting phase. Here we theoretically consider various types of N/S point contacts, away from the ballistic regime, and show that even when the superconductor under investigation is simple conventional in nature, depending on the shape, size and anatomy of the point contacts, a wide variety of spectral features may appear in the conductance spectra. Such features may misleadingly mimic theoretically expected signatures of exotic physical phenomena like Klein tunneling in topological superconductors, Andreev bound states in unconventional superconductors, multiband superconductivity and Majorana zero modes.

cond-mat.supr-con↗

arXiv2020

The pressure-enhanced superconducting phase of Sr$_x$-Bi$_2$Se$_3$ probed by hard point contact spectroscopy

The superconducting systems emerging from topological insulators upon metal ion intercalation or application of high pressure are ideal for investigation of possible topological superconductivity. In this context, Sr-intercalated Bi$_2$Se$_3$ is specially interesting because it displays pressure induced re-entrant superconductivity where the high pressure phase shows almost two times higher $T_c$ than the ambient superconducting phase ( $T_C\sim$ 2.9 K). Interestingly, unlike the ambient phase, the pressure-induced superconducting phase shows strong indication of unconventional superconductivity. However, since the pressure-induced phase remains inaccessible to spectroscopic techniques, the detailed study of the phase remained an unattained goal. Here we show that the high-pressure phase can be realized under a mesoscopic point contact, where transport spectroscopy can be used to probe the spectroscopic properties of the pressure-induced phase. We find that the point contact junctions on the high-pressure phase show unusual response to magnetic field supporting the possibility of unconventional superconductivity.

cond-mat.supr-con↗

arXiv2020

Developing a Multilingual Annotated Corpus of Misogyny and Aggression

In this paper, we discuss the development of a multilingual annotated corpus of misogyny and aggression in Indian English, Hindi, and Indian Bangla as part of a project on studying and automatically identifying misogyny and communalism on social media (the ComMA Project). The dataset is collected from comments on YouTube videos and currently contains a total of over 20,000 comments. The comments are annotated at two levels - aggression (overtly aggressive, covertly aggressive, and non-aggressive) and misogyny (gendered and non-gendered). We describe the process of data collection, the tagset used for annotation, and issues and challenges faced during the process of annotation. Finally, we discuss the results of the baseline experiments conducted to develop a classifier for misogyny in the three languages.

cs.CL↗

arXiv2020

High Entropy Alloy Nanoparticles Decorated, p-type 2D-Molybdenum Disulphide (MoS2) and Gold Schottky Junction Enhanced Hydrogen Sensing

Molybdenum Disulphide (MoS2) is an interesting material which exists in atomically thin multilayers and can be exfoliated into monolayer or a few layers for multiple applications. It has emerged as a promising material for development of such efficient sensors. Here, we have exfoliated and decorated MoS2 flakes with novel, single phase High Entropy Alloy (HEA) nanoparticles using facile and scalable cryomilling technique, followed by sonochemical method. It is found that the decoration of HEA nanoparticles impart surface enhanced Raman scattering effect and reduction in the work function of the material from 4.9 to 4.75 eV as measured by UV photoelectron spectroscopy. This change in the work function results in a schottky barrier between the gold contact and HEA decorated MoS2 flakes as a result of drastic changes in the surface chemical non-stoichiometry. The response to hydrogen gas is studied at temperatures 30 to 100 oC and it shows unusual p-type nature due to surface adsorbed oxygen species. The nanoscale junction formed between HEA and MoS2 shows 10 times increase in the response towards hydrogen gas at 80 oC. The experimental observations are further explained with DFT simulation showing more favourable hydrogen adsorption on HEA decorated MoS2 resulting for enhanced response.

cond-mat.mtrl-sci↗

arXiv2020

Tale of tails using rule augmented sequence labeling for event extraction

The problem of event extraction is a relatively difficult task for low resource languages due to the non-availability of sufficient annotated data. Moreover, the task becomes complex for tail (rarely occurring) labels wherein extremely less data is available. In this paper, we present a new dataset (InDEE-2019) in the disaster domain for multiple Indic languages, collected from news websites. Using this dataset, we evaluate several rule-based mechanisms to augment deep learning based models. We formulate our problem of event extraction as a sequence labeling task and perform extensive experiments to study and understand the effectiveness of different approaches. We further show that tail labels can be easily incorporated by creating new rules without the requirement of large annotated data.

cs.IR↗

arXiv2019

Spin-orbit driven spin depolarization in the ferromagnetic Weyl semimetal Co3Sn2S2

Co$_3$Sn$_2$S$_2$ has recently emerged as a ferromagnetic Weyl semi-metal. Theoretical investigation of the spin-split bands predicted half metalicity in the compound. Here, we report the detection of a spin polarized supercurrent through a Nb/Co$_3$Sn$_2$S$_2$ point contact where Andreev reflection is seen to be large indicating a large deviation from half metallicity. In fact, analysis of the Andreev reflection spectra reveals only 50\% spin polarization at the Fermi level of Co$_3$Sn$_2$S$_2$. Our theoretical calculations of electronic Density of States (DOS) reveal a spin depolarizing effect near the Fermi energy when the role of spin-orbit coupling is included. Inclusion of spin-orbit coupling also reveals particle-hole asymmetry that explains a large asymmetry observed in our experimental Andreev reflection spectra.

cond-mat.supr-con↗

arXiv2019

Alzheimer's Disease Brain MRI Classification: Challenges and Insights

In recent years, many papers have reported state-of-the-art performance on Alzheimer's Disease classification with MRI scans from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset using convolutional neural networks. However, we discover that when we split that data into training and testing sets at the subject level, we are not able to obtain similar performance, bringing the validity of many of the previous studies into question. Furthermore, we point out that previous works use different subsets of the ADNI data, making comparison across similar works tricky. In this study, we present the results of three splitting methods, discuss the motivations behind their validity, and report our results using all of the available subjects.

eess.IV↗

arXiv2019

SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval)

We present the results and the main findings of SemEval-2019 Task 6 on Identifying and Categorizing Offensive Language in Social Media (OffensEval). The task was based on a new dataset, the Offensive Language Identification Dataset (OLID), which contains over 14,000 English tweets. It featured three sub-tasks. In sub-task A, the goal was to discriminate between offensive and non-offensive posts. In sub-task B, the focus was on the type of offensive content in the post. Finally, in sub-task C, systems had to detect the target of the offensive posts. OffensEval attracted a large number of participants and it was one of the most popular tasks in SemEval-2019. In total, about 800 teams signed up to participate in the task, and 115 of them submitted results, which we present and analyze in this report.

cs.CL↗

arXiv2019

Predicting the Type and Target of Offensive Posts in Social Media

As offensive content has become pervasive in social media, there has been much research in identifying potentially offensive messages. However, previous work on this topic did not consider the problem as a whole, but rather focused on detecting very specific types of offensive content, e.g., hate speech, cyberbulling, or cyber-aggression. In contrast, here we target several different kinds of offensive content. In particular, we model the task hierarchically, identifying the type and the target of offensive messages in social media. For this purpose, we complied the Offensive Language Identification Dataset (OLID), a new dataset with tweets annotated for offensive content using a fine-grained three-layer annotation scheme, which we make publicly available. We discuss the main similarities and differences between OLID and pre-existing datasets for hate speech identification, aggression detection, and similar tasks. We further experiment with and we compare the performance of different machine learning models on OLID.

cs.CL↗

arXiv2019

A Comprehensive Study of Alzheimer's Disease Classification Using Convolutional Neural Networks

A plethora of deep learning models have been developed for the task of Alzheimer's disease classification from brain MRI scans. Many of these models report high performance, achieving three-class classification accuracy of up to 95%. However, it is common for these studies to draw performance comparisons between models that are trained on different subsets of a dataset or use varying imaging preprocessing techniques, making it difficult to objectively assess model performance. Furthermore, many of these works do not provide details such as hyperparameters, the specific MRI scans used, or their source code, making it difficult to replicate their experiments. To address these concerns, we present a comprehensive study of some of the deep learning methods and architectures on the full set of images available from ADNI. We find that, (1) classification using 3D models gives an improvement of 1% in our setup, at the cost of significantly longer training time and more computation power, (2) with our dataset, pre-training yields minimal ($<0.5\%$) improvement in model performance, (3) most popular convolutional neural network models yield similar performance when compared to each other. Lastly, we briefly compare the effects of two image preprocessing programs: FreeSurfer and Clinica, and find that the spatially normalized and segmented outputs from Clinica increased the accuracy of model prediction from 63% to 89% when compared to FreeSurfer images.

cs.CV↗

arXiv2018

Topological phases in hydrogenated gallenene and in its group elements

Nontrivial topology of Dirac (DSMs), Weyl (WSMs) and nodal line semimetals (NLSMs) are delineated by the novel band crossings near the Fermi level in the bulk and the appearance of exotic surface states. Among them, nodal line semimetals have gained immense interest due to the formation of one-dimensional nodal ring near the Fermi level. Using density funtional theory (DFT) calculations, we report that two dimensional (2D) NLSM phase can be hosted on hydrogen passivated (010) surface of gallium (gallenene) and on other group 13 elements, without inclusion of spin-orbit coupling (SOC). NLSM in these 2D systems is protected by the presence of crystalline (CS) along with inversion (IS) and the time reversal symmetry (TRS). In the presence of SOC, aluminane preserved its topological NLSM phase while in other single layered group 13 elements, a gap opened at the nodal point due to relatively stronger SOC effect. On applying tensile strain along with the inclusion of SOC, hydrogen passivated gallenene (gallenane) evolves into quantum spin Hall insulator with an indirect bandgap of 28 meV. The appearance of long range dissipationless linearly dispersive helical edge states and a large gap in gallenane make it promising for room temperature spintronics applications.

cond-mat.mtrl-sci↗

arXiv2018

Aggression-annotated Corpus of Hindi-English Code-mixed Data

As the interaction over the web has increased, incidents of aggression and related events like trolling, cyberbullying, flaming, hate speech, etc. too have increased manifold across the globe. While most of these behaviour like bullying or hate speech have predated the Internet, the reach and extent of the Internet has given these an unprecedented power and influence to affect the lives of billions of people. So it is of utmost significance and importance that some preventive measures be taken to provide safeguard to the people using the web such that the web remains a viable medium of communication and connection, in general. In this paper, we discuss the development of an aggression tagset and an annotated corpus of Hindi-English code-mixed data from two of the most popular social networking and social media platforms in India, Twitter and Facebook. The corpus is annotated using a hierarchical tagset of 3 top-level tags and 10 level 2 tags. The final dataset contains approximately 18k tweets and 21k facebook comments and is being released for further research in the field.

cs.CL↗

arXiv2018

Automatic Identification of Closely-related Indian Languages: Resources and Experiments

In this paper, we discuss an attempt to develop an automatic language identification system for 5 closely-related Indo-Aryan languages of India, Awadhi, Bhojpuri, Braj, Hindi and Magahi. We have compiled a comparable corpora of varying length for these languages from various resources. We discuss the method of creation of these corpora in detail. Using these corpora, a language identification system was developed, which currently gives state of the art accuracy of 96.48\%. We also used these corpora to study the similarity between the 5 languages at the lexical level, which is the first data-based study of the extent of closeness of these languages.

cs.CL↗

arXiv2017

Cooperative control of multi-agent systems to locate source of an odor

This work targets the problem of odor source localization by multi-agent systems. A hierarchical cooperative control has been put forward to solve the problem of locating source of an odor by driving the agents in consensus when at least one agent obtains information about location of the source. Synthesis of the proposed controller has been carried out in a hierarchical manner of group decision making, path planning and control. Decision making utilizes information of the agents using conventional Particle Swarm Algorithm and information of the movement of filaments to predict the location of the odor source. The predicted source location in the decision level is then utilized to map a trajectory and pass that information to the control level. The distributed control layer uses sliding mode controllers known for their inherent robustness and the ability to reject matched disturbances completely. Two cases of movement of agents towards the source, i.e., under consensus and formation have been discussed herein. Finally, numerical simulations demonstrate the efficacy of the proposed hierarchical distributed control.

eess.SY↗

arXiv2017

Giant Enhancement of Superconductivity in Zr Point Contacts

For certain complex superconducting systems, the superconducting properties get enhanced under mesoscopic point contacts made of elemental non-superconducting metals. However, understanding of the mechanism through which such contact induced local enhancement of superconductivity happens has been limited due to the complex nature of such compounds. In this paper we present giant enhancement of superconducting transition temperature (T$_c$) and superconducting energy gap ($Δ$) in a simple elemental superconductor Zr. While bulk Zr shows a critical temperature around 0.6\,K, superconductivity survives at Ag/Zr and Pt/Zr point contacts up to 3\,K with a corresponding five-fold enhancement of $Δ$. From first principles calculations we show that the enhancement in superconducting properties can be attributed to a modification in the electron-phonon coupling accompanied by an enhancement of the density of states which involves the appearance of a new electron band at the Ag/Zr interfaces.

cond-mat.supr-con↗