Saikat

Md Saikat Islam Khan

PhD Student in Computer Science

Hi! My name is Md Saikat Islam Khan, and I am a third-year PhD student in the BRAINS Lab at Rensselaer Polytechnic Institute. My advisor is Dr. Oshani Seneviratne.

Research Interests

I am a PhD student in the BRAINS Lab at Rensselaer Polytechnic Institute, where my research focuses on advancing distributed learning and explainability. I develop methods for privacy-preserving and robust training across vertically and horizontally partitioned data, alongside model-agnostic frameworks that generate sparse, actionable counterfactual explanations.

Currently, I am extending this work to complex question answering over distributed knowledge graphs. We aim to support multi-hop reasoning and query answering in federated settings without exposing raw entities, relations, or embeddings.

More broadly, my research analyzes and reconciles the trade-offs between privacy, robustness, interpretability, and utility in realistic deployment scenarios, supported by rigorous mathematical guarantees.

Education

Rensselaer Polytechnic Institute — Troy, New York, USA
Ph.D. in Computer Science  |  CGPA: 3.76  |  Aug 2023 – May 2028
Rensselaer Polytechnic Institute — Troy, New York, USA
M.Sc. in Computer Science  |  CGPA: 4.00  |  Aug 2023 – Dec 2025
Mawlana Bhashani Science and Technology University — Tangail, Bangladesh
B.Sc. in Computer Science  |  CGPA: 3.80  |  Jan 2015 – Dec 2019

Selected Publications

Fed-RD: Privacy-preserving federated learning for financial crime detection
Conference: IEEE Symposium on Computational Intelligence for Financial Engineering and Economics, 2024
Authors: Md. Saikat Islam Khan; Aparna Gupta; Oshani Seneviratne; Stacy Patterson | Paper |
A Differentially Private Blockchain-Based Approach for Vertical Federated Learning
Conference: IEEE International Conference on Decentralized Applications and Infrastructures, 2024
Authors: Linh Tran; Sanjay; Md. Saikat Islam Khan; Aaron; Stacy Patterson; Oshani Seneviratne | Paper |
Accurate brain tumor detection using deep convolutional neural network
Journal: Computational and Structural Biotechnology Journal, 2022
Authors: Md. Saikat Islam Khan; Anichur Rahman; Mostofa Kamal Nasir; Shahab S. Band; Amir Mosavi; Iman Dehzangi | Paper |
MultiNet: A deep neural network approach for detecting breast cancer through multi-scale feature fusion
Journal: Journal of King Saud University - Computer and Information Sciences, 2022
Authors: Md. Saikat Islam Khan; Nazrul Islam; Jia Uddin; Sifatul Islam; Mostofa Kamal Nasir | Paper |
Water quality prediction and classification based on principal component regression and gradient boosting classifier approach
Journal: Journal of King Saud University - Computer and Information Sciences, 2022
Authors: Md. Saikat Islam Khan; Ashef Shahrior; Razaul Karim; Mahmodul Hasan; Anichur Rahman | Paper |

Under Review

Fed-RD: Privacy-Preserving Federated Learning for Relational Data
Conference: IEEE Transactions on Big Data, 2026
Authors: MD Saikat Islam khan; Stacy Patterson; Oshani Seneviratne
SPICE: Sparse and Proximate Counterfactual Explanations via Feature-Importance-Weighted Perturbation.
Conference: Knowledge Discovery in Databases, 2026
Authors: MD Saikat Islam khan; Stacy Patterson; Oshani Seneviratne
FedKGQA: Privacy-Preserving Federated Learning for Multi-Hop Knowledge Graph Question Answering.
Conference: Knowledge Discovery in Databases, 2026
Authors: MD Saikat Islam khan; Oshani Seneviratne

Previous Experience

Lecturer, Bangladesh
Dhaka International University
Jan 2022–Jul 2023
Delivered undergraduate lectures in Computer Science, developed and evaluated course materials, supervised student projects, and guided research initiatives in emerging technologies.
Research Assistant - Bangladesh
Mawlana Bhashani Science and Technology University
Jan 2020–Dec 2021
Conducted research on machine learning algorithms and data analysis. Collaborated with senior researchers on cutting-edge AI projects. Mentors: Dr. Mostofa Kamal Nasir.

Skills

Programming Languages

  • Python
  • Java
  • Haskell
  • C++

Machine Learning

  • PyTorch
  • TensorFlow
  • Scikit-learn
  • Keras

Tools & Technologies

  • Git
  • Docker
  • AWS
  • Linux

Recent News

February 2026: Submitted my work on Counterfactual Explainability at KDD.
February 2026: Submitted my work on KGQA Explainability at KDD
April 2025: Passed my PhD Research Qualifying Exam.
January 2025: Filed a patent for Fed-RD (under review).
October 2024: Work in Fed-RD is accepted at CIFEr Read more
August 2024: Work in Private Blockchain VFL is accepted at IEEE DAPPS. Read more
Aug 2023: Joined RPI as a PhD student.

Reviewer Experience

  • KDD Conference 2026
  • AAAI Conference on Artificial Intelligence 2025
  • Informatics in Medicine Unlocked 2023
  • IEEE Access 2023