Publications And Tutorials

. Boosting Supervised Neural Relation Extraction with Distant Supervision. Electronic Thesis or Dissertation. Ohio State University, 2018., 2018.

PDF

. Tutorial - A Convolutional Encoder Model for Neural Machine Translation. NIPS Highlights (MLTrain), Learn How to code a paper with state of the art frameworks, 2017.

Code Poster Tutorial Workshop Azure Notebook

Experience

  • Research Assistant, The Ohio State University

    Currently working under the supervision of Prof. Huan Sun on developing techniques for effectively utilizing clean and noisy data in natural language processing.

  • Software Engineering Intern, AWS Deep Learning

    As part of the team that built Amazon Comprehend I worked on the following tasks for deployment of an NLP service (currently not part of the Comprehend suite) on AWS Infrastructure:

    • Performing extensive experiments using multiple machine learning models and datasets for the given task.
    • Creating end to end Serverless application using various AWS services like Step Functions, Lambda, Elastic Container Service, etc.
  • Research Assistant, Ubiquitous Knowledge Processing Lab, TU Darmstadt

  • Research Associate, Precog Research Group, IIIT Delhi, India

  • Software Engineer, Search Team, Infoedge India

  • Software Engineering Intern, Samsung Research India

  • Freelance Software Engineer, FunnelMailApp

Talks

Employing External Rich Knowledge for Machine Comprehension
Nov 28, 2016
Tables as Semi-structured Knowledge for Question Answering
Oct 17, 2016

Projects

Knowledge Distillation (using Tensorflow)

Training a smaller neural model using soft targets obtained from a larger model

Event & Participant Extraction from News Articles (Automatic Timeline Generation of News Events)

Extract events in News articles and their participants using a conditional random field classifier and extensive feature engineering

Fake News Challenge: Stance Detection for combating Fake News

Detecting the stance of the body of an article with respect to its title. Our team (OSUfnc) ranked 7th of the 50 teams that participated

Teaching

Contact