Sagnik Majumder

♂ - 25 years  (+1)-5129034773  [email protected]  Webpage  GitHub

Education
May 2025 University of Texas at Austin (UT)-Austin, Texas, United States
Doctor of Philosophy and Master of Science in Computer Science; GPA: 3.96/4.0

July 2018 Birla Institute of Technology and Science (BITS)-Pilani, Pilani, Rajasthan, India
Bachelor of Engineering (Hons.) in Electronics and Instrumentation
Thesis: “Neural Architecture Meta-learning via Reinforcement” | Advisor: Prof. V. Ramesh
GPA: 9.55/10, Distinction and ranked 2nd out of 100 students

Internships and Research

Jan 2020 - present Research assistant at UT Austin Vision Lab
Advisor - Prof. Kristen Grauman
• Introduced the new task of active audio-visual source separation and proposed
a novel deep RL based method to solve it
• Built an RL based hierarchical audio-visual navigation system that combined a
novel end-to-end waypoint prediction model and an geometric motion plan-
ner, and also leveraged a novel acoustic map design

Aug 20219 - Dec 2019 Student in Graduate Natural Language Processing course at UT Austin
Advisor - Prof. Greg Durrett
• Built a novel state-of-the-art adversarial defense for question answering that
uses a model-agnostic answer reranking mechanism by computing named en-
tity overlap between questions and candidate answers

Jan 2018 - May 2019 Research assistant at Goethe University

Advisor - Prof. Visvanathan Ramesh
• Built a continual learning framework by integrating a variational autoencoder
based deep generative replay model and a statistical outlier rejection tech-
nique (OpenSet) that outperformed the state-of-the-art
• Curated a novel concrete defect dataset; meta-learned task specific neural
architectures that outperformed strong baselines and transfer-learned models

May 2017 - May 2019 Research intern at Frankfurt Institute for Advanced Studies
Advisor - Prof. Christoph Malsburg
• Created a distortion invariant handwritten digit recognition system with Ga-
bor filters and an elastic graph matching algorithm
• Worked on motion parameter estimation and prediction of rigid rotating ob-
jects and implemented a neural version of the Kalman filter

1
Peer-reviewed Publications and Submissions
• Sagnik Majumder, Ziad Al-Halah, Kristen Grauman. ”Move2Hear: Active Audio-Visual Source Sepa-
ration”. Under Review. [arXiv], [Project Page]
• Changan Chen, Sagnik Majumder, Ziad Al-Halah, Ruohan Gao, Santhosh K. Ramakrishnan, Kristen
Grauman. ”Learning to Set Waypoints for Audio-Visual Navigation”. ICLR 2021.
[Publication], [Project Page]
• Sagnik Majumder, Chinmoy Samant, Greg Durrett. ”Model Agnostic Answer Reranking System for
Adversarial Question Answering”. EACL 2021 Student Research Workshop. [Publication].
• Martin Mundt, Sagnik Majumder, Iuliia Pliushch, Yong Won Hong, Visvanathan Ramesh. ”Unified
Probabilistic Deep Continual Learning through Generative Replay and Open Set Recognition”.
Under review. [Preprint], [Codebase]
• Martin Mundt, Sagnik Majumder, Sreenivas Narasimha Murali, Panagiotis Panetsos, Visvanathan
Ramesh. ”Meta-learning Convolutional Neural Architectures for Multi-target Concrete Defect Classi-
fication with the COncrete DEfect BRidge IMage Dataset”. CVPR 2019.
[Main body], [Supplementary], [Codebase]
• Martin Mundt, Sagnik Majumder, Tobias Weis, Visvanathan Ramesh. ”Rethinking Layer-wise Fea-
ture Amounts in Convolutional Neural Network Architectures”. NeurIPS 2018 Workshop: Critiquing
and Correcting Trends in Machine Learning.
[Workshop web-page with link to publication], [Publication], [Codebase]
• Martin Mundt, Iuliia Pliushch, Sagnik Majumder, Visvanathan Ramesh. ”Open Set Recognition
Through Deep Neural Network Uncertainty: Does Out-of-Distribution Detection Require Generative
Classifiers?”. ICCV 2019 Workshop: Statistical Deep Learning for Computer Vision (SDLCV).
[Publication]
• Sagnik Majumder, C. von der Malsburg, Aashish Richhariya, Surekha Bhanot, “Handwritten Digit
Recognition by Elastic Matching” Journal of Computers vol. 13, no. 9, pp. 1067-1074, 2018.
[Publication], [Codebase]
• Rishabh Bhardwaj, Sagnik Majumder, Pawan K. Ajmera, Soumendu Sinha, Rishi Sharma, R. Mukhiya,
Pratik Narang.“Temperature compensation of ISFET based pH sensor using artificial neural networks”
. In: Micro and Nanoelectronics (RSM), 2017 IEEE Regional Symposium on. IEEE. 2017, pp. 155–158.
[Publication]
• Rishabh Bhardwaj, Soumendu Sinha, Nishad Sahu, Sagnik Majumder, Pratik Narang, Ravindra
Mukhiya. ”Modeling and Simulation of Temperature Drift for ISFET based pH Sensor and its Com-
pensation through Machine Learning Techniques ”. International Journal of Circuit Theory and
Applications 2019.
[Publication]

Coursework
Graduate: Deep Learning Seminar; Reinforcement Learning: Theory & Practice;
Robot Learning; Natural Language Processing; Math in Deep Learning;
Statistical Models for Health and Behavioral Sciences;
Algorithms: Techniques and Theory; Programming Languages;
Undergraduate: Neural Networks & Fuzzy Logic; Machine Learning; Advanced Calculus;
Linear Algebra and Complex Variables; Probability and Statistics;
Computer Programming; Operating Systems; Object Oriented Programming;
Advanced Computer Architecture; Algorithms and Complexity; Data Structures;
Discrete Mathematics
MOOC: Stanford’s CS231n: Convolutional Neural Networks for Visual Recognition;
Stanford’s CS224n:Natural Language Processing with Deep Learning;
UC Berkeley’s CS294: Deep Reinforcement Learning

2
Professional Service
Reviewer: RA-L ’21; ICRA ’21

Software Skills
Programming Language: Python; C; C++; Java; Matlab
Autodifferentiation Framework: PyTorch; Tensorflow; Caffe
Python Package: Numpy; Scipy; SK-learn; Matplotlib; Seaborn; Plotly
Operating System: Linux (Debian, Ubuntu); MS Windows
Distributed Version Control: Git
Document Preparation: LATEX; MS Word

Academic Honors and Achievements

November 2020 TOEFL iBT: 118 (Reading: 29, Listening: 30 Speaking: 29, Writing: 30)
July 2018 GRE: 334 (Quantitative: 170, Verbal: 164, AWA: 5.0)
Jan 2016 - Jun 2018 Received merit scholarship for academic excellence from BITS Pilani for 5
consecutive semesters
March 2017 Secured 2nd place in paper presentation at APOGEE, BITS Pilani technical festival
December 2016 Received DAAD WISE scholarship 2017 for research internship in Germany
June 2014 Ranked in top 0.50% in IIT-JEE and 64 in WBJEE
February 2014 Offered KVPY fellowship by the Department of Science and Technology, Govt.
of India

Teaching Experience
Semester 1, 2017-18: Teaching assistant for ”Neural Networks and Fuzzy Logic” at BITS Pilani

Co-curricular Activities
2016-17: Project coordinator of Instrumentation Forum, BITS Pilani
2014-17: Member of BITS Firefox Community, Google Developers’ Group and
Instrumentation Forum at BITS PIlani

Extra-curricular Activities
2016-17: Cultural secretary of Moruchhaya, the Bengali cultural association at BITS Pilani
2014-18: Member of Moruchhaya