CV

Education

• Ph.D. in Computer Science and Engineering, Indian Institute of Technology Indore, (Aug. 2021 - present)
• M.S.(Research) in Computer Science and Engineering, Indian Institute of Technology Indore, (Aug. 2019 - Jul. 2021)
• B.Tech in Computer Science and Engineering, Dr. A.P.J. Abdul Kalam Technical University, Lucknow, (Aug. 2013 - Jul. 2017)

Projects

• Adversarial Attacks on Medical Image Segmentation Models, (Aug. 2021 - present)
  • Devised a novel attack architecture that selects the best surrogate loss function to perform attacks on medical image segmentation models while adding minimum adversarial perturbations.
  • The architecture was able to fool the state-of-the-art medical image segmentation models with an attack success rate of more than 99%.
• Defence against Adversarial Attacks on NLP Models, (Jan. 2021 - Jun. 2021)
  • Devised two novel transformation based defences that can circumvent state-of-the-art attacks.
  • The proposed defence overcome the drawbacks of the existing spelling correction based defences.
• Adversarial Attacks on Audio-Visual Speech Recognition Models, (May. 2020 - Nov. 2020)
  • Devised a novel attack to breach the state-of-the-art AVSR models by simultaneously deceiving the detection algorithm, with an attack success rate of 100%.
• Adversarial Attacks on Visual Speech Recognition Models, (Dec. 2019 - Apr. 2020)
  • Devised a novel algorithm to perform an attack on the state-of-the-art VSR models with a 100% attack success rate.
  • The algorithm was able to circumvent popular defences with an attack success rate of more than 99%.

Publications

FATALRead - Fooling visual speech recognition models

DARE: Deceiving Audio-Visual Speech Recognition model

Contextual Emotion Learning Challenge

TRIESTE: Translation based defense for text classifiers

D-NEXUS: Defending text networks using summarization

RADIANT: Better rPPG estimation using signal embeddings and Transformer

ALPINE: Improving Remote Heart Rate Estimation using Contrastive Learning

Work experience

• Kwench Global Technologies Pvt. Ltd. (Noida, Uttar Pradesh, Oct. 2017 - Apr. 2018)
  • Software Engineer
  • Developed Workplace Bots in a team that is Product Partner of Workplace, an enterprise connectivity platform by Facebook.
  • Worked on the Reward and Recognition Platform that is trusted by over 500 companies across 63 countries.
  • Developed rules for the Reward Platform, which helps provide the clients with numerous configurations of the Reward System of their choice with no change in the code.
• Novoinvent Softwares Pvt. Ltd. (Noida, Uttar Pradesh, Jul. 2017 - Sep. 2017)
  • Software Engineer
  • Worked on a US government insurance project for the state of Maryland, “Maryland Health Connection” which is the state’s official health insurance marketplace for individuals and families to compare and enrol in health insurance.
• Software Development Centre - Software Incubator. (Ghaziabad, Uttar Pradesh, Aug. 2014 - May 2017)
  • Student Co-ordinator/ Software Engineer
  • Developed a Biometric Attendance System for the faculty and staff of Ajay Kumar Garg Engineering College (AKGEC), Ghaziabad.
  • Developed and maintained more than 10 portals, websites and platforms related to AKGEC.
  • Web Team Member, Event Organiser: Tech Trishna’15, the annual national technical festival of AKGEC, which saw a footfall of 1500+ participants.
  • Developed an online registration and slot booking portal in the Blood Donation Camp, 2016 & 2017.
• My Own Space Research Consultancy. (Ghaziabad, Uttar Pradesh, Jun. 2016 - Jul. 2016)
  • Student Intern
  • Designed and developed a website that provides an easy solution to buyers for accessing reviews for any property and helps them make the correct decision.
• Sarvodya Ventures. (Ghaziabad, Uttar Pradesh, Jun. 2015)
  • Student Intern
  • Worked on a live portal that connects transporters, truck drivers, customers and other related entities.