Papers by Vishwajeet Kumar

Health Policy and Planning
India has made significant progress in improving maternal and child health. However, there are pe... more India has made significant progress in improving maternal and child health. However, there are persistent disparities in maternal and child morbidity and mortality in many communities. Mistreatment of women in childbirth and gender-based violence are common and reduce women’s sense of safety. Recently, the Government of India committed to establishing a specialized midwifery cadre: Nurse Practitioners in Midwifery (NPMs). Integration of NPMs into the current health system has the potential to increase respectful maternity care, reduce unnecessary interventions, and improve resource allocation, ultimately improving maternal–newborn outcomes. To synthesize the evidence on effective midwifery integration, we conducted a desk review of peer-reviewed articles, reports and regulatory documents describing models of practice, organization of health services and lessons learned from other countries. We also interviewed key informants in India who described the current state of the healthcare...

Physical programming and conjoint analysis-based redundancy allocation in multistate systems: A Taguchi embedded algorithm selection and control (TAS&C) approach
Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, 2009
Amidst increasing system complexity and technological advancements, the manufacturer aims to win ... more Amidst increasing system complexity and technological advancements, the manufacturer aims to win the consumer's trust to maintain his or her permanent goodwill. This expectation directs the manufacturer to address the problem of attaining desired quality and reliability standards; hence, the measure of performance of a system in terms of reliability and utility optimization poses an issue of primary concern. In order to meet the requirement of a reliable and trouble-free product, optimal allocation of all conflicting parameters is essential during the design phase of a system. With this in mind, this paper presents a physical programming and conjoint analysis-based redundancy allocation model (PPCA-RAM) for a multistate series—parallel system. Use of physical programming approach is the key feature of the proposed algorithm to eliminate the need for multi-objective optimization. Physical programming methodology provides an adequate balance among various associated performance me...

Proceedings of the National Academy of Sciences, 2010
Silica (SiO 2 ) is an abundant component of the Earth whose crystalline polymorphs play key roles... more Silica (SiO 2 ) is an abundant component of the Earth whose crystalline polymorphs play key roles in its structure and dynamics. First principle density functional theory (DFT) methods have often been used to accurately predict properties of silicates, but fundamental failures occur. Such failures occur even in silica, the simplest silicate, and understanding pure silica is a prerequisite to understanding the rocky part of the Earth. Here, we study silica with quantum Monte Carlo (QMC), which until now was not computationally possible for such complex materials, and find that QMC overcomes the failures of DFT. QMC is a benchmark method that does not rely on density functionals but rather explicitly treats the electrons and their interactions via a stochastic solution of Schrödinger’s equation. Using ground-state QMC plus phonons within the quasiharmonic approximation of density functional perturbation theory, we obtain the thermal pressure and equations of state of silica phases up ...
Incremental costs of scaling up kangaroo mother care: Results from implementation research in Ethiopia and India
Acta Paediatrica

Cornell University - arXiv, Dec 14, 2021
Question answering (QA) over tables and text linked from table elements, also called Text-TableQA... more Question answering (QA) over tables and text linked from table elements, also called Text-TableQA, has witnessed significant research in recent years, because tables are often found embedded in documents along with related text. HybridQA and OTT-QA are the two bestknown TextTableQA datasets, with questions that are best answered by combining information from both table cells and linked text passages. A common challenge in both datasets, and TextTableQA in general, is that the training instances include just the question and answer, where the gold answer may match not only multiple table rows but also multiple text spans within the scope of a row and its associated text. This leads to a noisy multiinstance training regime. We present MITQA, a transformer-based TextTableQA system that is explicitly designed to cope with distant supervision along both these axes, through a multi-instance loss objective, together with careful curriculum design. Our experiments show that the proposed multi-instance distant supervision approach helps MITQA get much better EM and F1 scores than existing baselines for both HybridQA and OTT-QA, with MITQA currently being at the top of Hy-bridQA leaderboard with a held out test set.
Hirschhorn_et_al-2018-International_Journal_of_Gynecology_&_Obstetrics.pdf
Publication- for reference
OAMS_DATASET_FINAL.SAS7BDAT
Effective Integration of the Opportunity‐Ability‐Motivation behavior change framework into a coaching‐based WHO Safe Childbirth Checklist program in India
This dataset contains data referenced in the publication, "Effective Integration of the Oppo... more This dataset contains data referenced in the publication, "Effective Integration of the Opportunity‐Ability‐Motivation behavior change framework into a coaching‐based WHO Safe Childbirth Checklist program in India".
CALL_CENTER_EFFICIENCY.SAS7BDAT
Additional file 3: Figure S2. of Effectiveness of the WHO Safe Childbirth Checklist program in reducing severe maternal, fetal, and newborn harm in Uttar Pradesh, India: study protocol for a matched-pair, cluster-randomized controlled trial
BetterBirth trial SPIRIT figure. (DOCX 32 kb)
Additional file 1: Figure S1. of Effectiveness of the WHO Safe Childbirth Checklist program in reducing severe maternal, fetal, and newborn harm in Uttar Pradesh, India: study protocol for a matched-pair, cluster-randomized controlled trial
BetterBirth Safe Childbirth Checklist: the WHO Safe Childbirth Checklist adapted for alignment wi... more BetterBirth Safe Childbirth Checklist: the WHO Safe Childbirth Checklist adapted for alignment with Indian national guidelines. (PDF 5557 kb)
REPOSITORY_BABYLEVEL_OUTCOMES.SAS7BDAT
MAIN_OUTCOMES_NEJM_FILES.ZIP
UNPACKING_THE_NULL_DATA.ZIP
Unpacking the Null: Facility-Level Response to a WHO Safe Childbirth Checklist Intervention in the BetterBirth Trial in Uttar Pradesh, India
This dataset contains the data referenced in the publication "Unpacking the Null: Facility-L... more This dataset contains the data referenced in the publication "Unpacking the Null: Facility-Level Response to a WHO Safe Childbirth Checklist Intervention in the BetterBirth Trial in Uttar Pradesh, India".
Mother Knows Best: The Practical Realities of Promoting Exclusive Breastfeeding in India
controlled trial in Uttar Pradesh, India
adaptive study design process prior to initiation of BetterBirth, a large-scale randomized

ArXiv, 2021
Solving math word problems (MWPs) is an important and challenging problem in natural language pro... more Solving math word problems (MWPs) is an important and challenging problem in natural language processing. Existing approaches to solve MWPs require full supervision in the form of intermediate equations. However, labeling every math word problem with its corresponding equations is a time-consuming and expensive task. In order to address this challenge of equation annotation, we propose a weakly supervised model for solving math word problems by requiring only the final answer as supervision. We approach this problem by first learning to generate the equation using the problem description and the final answer, which we then use to train a supervised MWP solver. We propose and compare various weakly supervised techniques to learn to generate equations directly from the problem description and answer. Through extensive experiment, we demonstrate that even without using equations for supervision, our approach achieves an accuracy of 56.0 on the standard Math23K dataset (Wang et al., 201...
Question Generation from Text and Multi-hop Question Generation and Answering from Knowledge Base
Question answering (QA) is the task of finding out the correct answer given a question and a know... more Question answering (QA) is the task of finding out the correct answer given a question and a knowledge source such as unstructured text or structured knowledge base (KB). In contrast, the task of question generation (QG) is a reverse task to generate a corresponding natural language question given knowledge in structured (KB) or unstructured form (text) and optionally a target answer. The motivation for QG is to generate large scale high-quality QA training data, which will help in improving the performance of QA model and also in increasing the efficiency of human annotators in QA dataset construction. In this thesis, we study the problem of automatically generating meaningful, relevant and challenging questions from sentences, paragraphs, and knowledge base.
Uploads
Papers by Vishwajeet Kumar