Elkhan Ismayilzada | Michigan State University

About Me

I am a PhD candidate in CSE at Michigan State University. My current research interests include 3D reconstruction, pose estimation, action recognition, object detection.

Research Interests

Computer Vision: pose estimation, 3D reconstruction, action recognition, object detection.
Machine Learning: deep learning, meta-learning, incremental learning.

Education

PhD in Computer Science and Engineering, Michigan State University, USA
2024 – Present
M.Sc. in Computer Science and Engineering, UNIST, South Korea
2022 – 2024
B.Sc. in Computer Science and Engineering, UNIST, South Korea
2018 – 2022

Research Experience

Graduate Research Assistant, Department of Computer Science and Engineering
UNIST Vision and Learning, Ulsan National Institute of Science and Technology
Project: Pose estimation
Supervisor: Seungryul Baek
Focus: Hands-object pose estimation
Sep 2022 – Present
Research Assistant, Department of Computer Science and Engineering
Statistical Decision Making Lab, Ulsan National Institute of Science and Technology
Project: Contextual Multi-armed bandit algorithms
Supervisor: Gi-Soo Kim
Focus: Semiparametric contextual bandits
Mar 2021 – Oct 2022
Research Assistant, Department of Computer Science and Engineering
Applied Cryptography Lab, Ulsan National Institute of Science and Technology
Project: Privacy-preserving Machine Learning
Supervisor: Miran Kim
Focus: Secure computation during training and inference phase of model using HE
Mar 2021 – Sep 2022
Research Assistant, Department of Computer Science and Engineering
Software Engineering Lab, Ulsan National Institute of Science and Technology
Project: Investigations on automated program repair tools
Supervisor: Jooyong Yi
Focus: Overfitting problem of automated program repair tools
Jan 2020 – Feb 2021

Work experience

Teaching Assistant, Department of Computer Science and Engineering
Organized the exams, made and evaluated programming assignments in CSE472 Computer Vision.
Sep 2022 - Dec 2022

Publications

Conferences

Transformer-based Unified Recognition of Two Hands Manipulating Objects
Hoseong Cho, Chanwoo Kim, Jihyeon Kim, Seongyeong Lee, Elkhan Ismayilzada, and Seungryul Baek
Proc. of IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2023
[Paper]

Journals

Poracle: Testing Patches Under Preservation Conditions to Combat the Overfitting Problem of Program Repair
Elkhan Ismayilzada, Md Mazba Ur Rahman, Dongsun Kim, Jooyong Yi
ACM Transactions on Software Engineering and Methodology (TOSEM), 2023
[Paper] [Code]
Secure human action recognition by encrypted neural network inference
Miran Kim, Xiaoqian Jiang, Kristin Lauter, Elkhan Ismayilzada, and Shayan Shams
Nature Communications 13 (1), 1-13, 2022
[Paper]
Speeding up Constraint-Based Program Repair Using a Search-Based Technique
Jooyong Yi, and Elkhan Ismayilzada
Information and Software Technology 146, 106865, 2022
[Paper] [Code]

Posters

Development of news article recommendation system via reinforcement learning
Elkhan Ismayilzada, Sayem K. Chowdury, and Mubarrat T. Chowdury
Poster presented at: UNIST; Dec 3 2021; Ulsan, Republic of Korea.
[Poster] [Code]
Bang for the Buck: Filtering Out Hundreds of Incorrect Patches with a Single Generalized Test
Elkhan Ismayilzada, Mazba Ur Rahman, Dongsun Kim, and Jooyong Yi
International Conference on Software Engineering (Submitted).
[Poster] [Code]

Projects

UNIST Projects
Coursework of all the courses at UNIST.
[Code]
Tapmaca
Simple quiz game which is made with Unity, published in Google Play Store and has over 100 thousand downloads.
[Google Play] [Code]
Paytaxt
Simple quiz game which is made with Unity, published in Google Play Store and has over 10 thousand downloads.
[Google Play] [Code]
Movie Recommendation App
Web App that is made with Spring for backend and jQuery / HTML for frontend.
[Code]
E-Commerce in React
Web App that is made with React + Redux. For storing user’s information, Firebase is integrated.
[Code]