Ritwick Chaudhry

hover over me @ amazon.com

hover over me @ gmail.com

Amazon Web Services

Biography

I am working as an Applied Scientist at Amazon (AWS) AI , focussing on developing novel technologies in Deep Learning and Computer Vision. I graduated from Carnegie Mellon University with a Masters' in Computer Science , and I have completed my B.Tech from Indian Institute of Technology (IIT) Bombay . I have also worked for a year at Adobe Research , and recently completed an internship at Uber Advanced Technologies Group (ATG) , building sophisticated joint perception/prediction models for making autonomous vehicles a reality. In the past, I have also completed research internships at Adobe Research and the Johns Hopkins Univsersity under Prof. Suchi Saria . I have published my work in top conferences like CVPR, AAAI, WACV and have authored 2 issued and 4 filed patents in the US Patent Office .


B.Tech, Computer Science IIT Bombay 2014 - 2018	Research Intern Johns Hopkins University Summer 2016	Research Intern Adobe Research Summer 2017	Research Engineer Adobe Research Jun 2018 - Jul 2019	MS in CS, Carnegie Mellon University Aug 2019 - Dec 2020	Applied Scientist AWS AI Feb 2021 - Present

Updates

[Oct 2021]

Our Patent on System to facilitate exchange of data segments between data aggregators and data consumers has been issued in the US Patent Office

[Jun 2021]

Invited to give a talk at the CVPR Workshop on Chart Question Answering. The recorded session is available online.

[Jun 2021]

Our work on Track, Check, Repeat: An EM Approach to Unsupervised Tracking was presented at CVPR 2021.

[Mar 2021]

Our patent on Personalized e-learning using a deep-learning based knowledge tracing and hint-taking propensity model has been issued in the US Patent Office

[Feb 2021]

Presented our work Scene Graph Embeddings using Relative Similarity Supervision at AAAI 2021.

[Feb 2021]

Started working as an Applied Scientist at Amazon Web Services (AWS) AI in the Bay Area.

[Dec 2020]

Graduated from Carnegie Mellon University with a Masters' in Computer Science, securing a GPA of 4.29 out of 4.00.

Publications

Scene Graph Embeddings Using Relative Similarity Supervision

Paridhi Maheshwari ^†, Ritwick Chaudhry ^†, Vishwa Vinay († denotes equal contribution)

AAAI 2021

Abstract PDF Talk BibTeX

@inproceedings{maheshwari2021scene,
  title={Scene Graph Embeddings Using Relative Similarity Supervision},
  author={Maheshwari, Paridhi and Chaudhry, Ritwick and Vinay, Vishwa},
  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
  volume={35},
  number={3},
  pages={2328--2336},
  year={2021}
}

Track, Check, Repeat: An EM Approach to Unsupervised Tracking

Adam W. Harley, Yiming Zuo, Jing Wen, Ayush Mangal, Shubhankar Potdar, Ritwick Chaudhry, Katerina Fragkiadaki,

CVPR 2021

Abstract PDF Video BibTeX

@InProceedings{Harley_2021_CVPR,
    author    = {Harley, Adam W. and Zuo, Yiming and Wen, Jing and Mangal, Ayush and Potdar, Shubhankar and Chaudhry, Ritwick and Fragkiadaki, Katerina},
    title     = {Track, Check, Repeat: An EM Approach to Unsupervised Tracking},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2021},
    pages     = {16581-16591}
}

LEAF-QA: Locate, Encode & Attend for Figure question answering

Ritwick Chaudhry, Sumit Shekhar, Utkarsh Gupta Pranav Maneriker, Prann Bansal, Ajay Joshi

WACV 2020, Invited Talk at CVPR 2021 Workshop on Chart Question Answering (CQAW)

Abstract PDF BibTeX Talk

@InProceedings{Chaudhry_2020_WACV,
author = {Chaudhry, Ritwick and Shekhar, Sumit and Gupta, Utkarsh and Maneriker, Pranav and Bansal, Prann and Joshi, Ajay},
title = {LEAF-QA: Locate, Encode & Attend for Figure Question Answering},
booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
month = {March},
year = {2020}
}

AB Initio Tomography With Object Heterogeneity and Unknown Viewing Parameters

Arunabh Ghosh^†, Ritwick Chaudhry^†, Ajit Rajwade († denotes equal contribution)

IEEE ICIP 2019

Abstract PDF BibTeX

@inproceedings{ghosh2019ab,
  title={AB Initio Tomography With Object Heterogeneity and Unknown Viewing Paramete},
  author={Ghosh, Arunabh and Chaudhry, Ritwick and Rajwade, Ajit},
  booktitle={2019 IEEE International Conference on Image Processing (ICIP)},
  pages={1257--1261},
  year={2019},
  organization={IEEE}
}

ICDAR 2019 Competition on Harvesting Raw Tables from Infographics (CHART-Infographics)

Kenny Davila, Bhargava Urala Kota, Srirangaraj Setlur, Venu Govindaraju, Christopher Tensmeyer, Sumit Shekhar Ritwick Chaudhry,

IEEE ICDAR 2019

Abstract PDF BibTeX

@inproceedings{davila2019icdar,
  title={ICDAR 2019 Competition on Harvesting Raw Tables from Infographics (CHART-Infographics)},
  author={Davila, Kenny and Kota, Bhargava Urala and Setlur, Srirangaraj and Govindaraju, Venu and Tensmeyer, Christopher and Shekhar, Sumit and Chaudhry, Ritwick},
  booktitle={2019 International Conference on Document Analysis and Recognition (ICDAR)},
  pages={1594--1599},
  year={2019},
  organization={IEEE}
}

Modeling Hint-Taking Behavior and Knowledge State of Students with Multi-Task Learning

Ritwick Chaudhry, Harvineet Singh, Pradeep Dogga, Shiv Kumar Saini,

MedialEval 2018

Abstract PDF BibTeX

Interactive learning environments facilitate learning by providing hints to fill the gaps in the understanding of a concept. Studies suggest that hints are not used optimally by learners. Either they are used unnecessarily or not used at all. It has been shown that learning outcomes can be improved by providing hints when needed. An effective hinttaking prediction model can be used by a learning environment to make adaptive decisions on whether to withhold or provide hints. Past work on student behavior modeling has focused extensively on the task of modeling a learner’s state of knowledge over time, referred to as knowledge tracing. The other aspects of a learner’s behavior such as tendency to use hints has garnered limited attention. Past knowledge tracing models either ignore the questions where a hint was taken or label hints taken as an incorrect response. We propose a multi-task memory-augmented deep learning model to jointly predict the hint-taking and the knowledge tracing task. The model incorporates the effect of past responses as well as hints taken on both the tasks. We apply the model on two datasets–ASSISTments 2009-10 skill builder dataset and Junyi Academy Math Practicing Log. The results show that deep learning models efficiently leverage the sequential information present in a learner’s responses. The proposed model significantly out-performs the past work on hint prediction by at least 12% points. Moreover, we demonstrate that jointly modeling the two tasks improves performance consistently across the tasks and the datasets, albeit by a small amount.

@article{chaudhry2018modeling,
  title={Modeling Hint-Taking Behavior and Knowledge State of Students with Multi-Task Learning.},
  author={Chaudhry, Ritwick and Singh, Harvineet and Dogga, Pradeep and Saini, Shiv Kumar},
  journal={International Educational Data Mining Society},
  year={2018},
  publisher={ERIC}
}

Show and Recall@ MediaEval 2018 ViMemNet: Predicting Video Memorability

Ritwick Chaudhry, Manoj Kilaru, Sumit Shekhar

Educational Data Mining 2018

Abstract PDF BibTeX

@article{chaudhry2018show,
  title={Show and Recall@ MediaEval 2018 ViMemNet: Predicting Video Memorability},
  author={Chaudhry, Ritwick and Kilaru, Manoj and Shekhar, Sumit},
  journal={MediaEval 2018},
  volume={1},
  pages={G1},
  year={2018}
}

Tomographic reconstruction using global statistical priors

Preeti Gopal, Ritwick Chaudhry, Sharat Chandran, Imants Svalbe, Ajit Rajwade

IEEE DICTA 2017

Abstract PDF BibTeX

@inproceedings{gopal2017tomographic,
  title={Tomographic reconstruction using global statistical priors},
  author={Gopal, Preeti and Chaudhry, Ritwick and Chandran, Sharat and Svalbe, Imants and Rajwade, Ajit},
  booktitle={2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA)},
  pages={1--8},
  organization={IEEE}
}

Patents

Personalized e-learning using a deep-learning based knowledge tracing and hint-taking propensity model

Shiv Kumar Saini, Ritwick Chaudhry, Pradeep Dogga, Harvineet Singh,

Patent Number: 10,943,497, Issued: 03/09/2021

System to facilitate exchange of data segments between data aggregators and data consumers

Shiv Kumar Saini, Ritwick Chaudhry, Harvineet Singh, Bhavya Bahl, Sriya Sainath, Savya Sindhu Gupta,

Patent Number: 11,151,532 Issued: 10/19/2021

Chart Question Answering

Sumit Shekhar, Ritwick Chaudhry, Utkarsh Gupta, Prann Bansal, Ajay Joshi

Patent Application Number: 16,749,044 Filed: 01/22/2021

Scene Graph Embeddings Using Relative Similarity Supervision

Paridhi Maheshwari, Ritwick Chaudhry, Vishwa Vinay

Patent Application Number: 17,337,801 Filed: 06/03/2021

Key-Value Memory Network for predicting time-series metrics of target entities

Ayush Chauhan, Shiv Kumar Saini, Parth Gupta, Archiki Prasad, Amireddy Prashanth Reddy Ritwick Chaudhry,

Patent Application Number: 17,337,801 Filed: 06/03/2021

A Method for Parsing and Reflowing Infographics using Structured Lists and Groups

Sumit Shekhar, Zoya Bylinski, Tushar Gurjar, Ritwick Chaudhry, Ayush Goyal,

Patent Application Number: 20,030,347 Filed:

Teaching Experience

Carnegie Mellon University

16720 - Computer Vision (Fall 2019), with Prof. Srinivasa Narasimhan
18661 - Intro to Machine Learning for Engineers (Spring 2019), with Prof. Gauri Joshi

IIT Bombay

CS207 - Discrete Structures (Spring 2017), with Prof. S. Akshay
CS 347/333 - Operating Systems (Fall 2017), with Prof. Puru Kulkarni
CS 228 - Login in Computer Science (Spring 2016), with Prof. Krishna S.
CS207 - Discrete Structures (Fall 2016), with Prof. S. Akshay
MA 108 - Differential Equations (Fall 2015), with Prof. Santanu Dey