Fatema Siddika

Fatema Siddika

Position
  • PhD Student
Fatema Siddika is a Ph.D. student in the Department of Computer Science at Iowa State University, working in the Software Analytics and Pervasive Parallelism (SwAPP) Lab under the supervision of Dr. Ali Jannesari. Fatema’s doctoral research focuses on Federated Learning (FL) in heterogeneous environments, emphasizing communication-efficient, privacy-preserving approaches for aligning diverse models and enhancing global learning. Her work addresses challenges such as minimizing information loss during server aggregation and reducing communication costs for LLMs in FL settings. She has also worked as a Givens Research Associate at Argonne National Laboratory, focusing on fine-tuning LLMs for diverse tasks, exploring sparsity patterns, and advancing model interpretability. She received her B.Sc. and M.Sc. degrees in Computer Science and Engineering from Jagannath University. She joined Iowa State in 2021 for her Ph.D., contributing to advanced research in Federated Learning and large-scale distributed systems.

Contact

Social Media and Websites

Area of Expertise

  • Federated Learning
  • FedLLM Finetuning
  • Representation Learning

Education

  • BSc., Computer Science and Engineering, Jagannath University, 2015
  • MSc, Computer Science and Engineering, Jagannath University, 2017

Publications

Siddika*, F., Hossen*, M. A. & Zhang, W. Fair bandwidth allocation at Edge servers for hierarchical, distributed, and concurrent Federated Learning Processes in The 10th International Conference on Fog and Mobile Edge Computing (FMEC 2025) (May 17, 2025).

Siddika, F., Hossen, M. A. & Saha, S. Transition from IPv4 to IPv6 in Bangladesh: The competent and enhanced way to follow in 2017 International Conference on Networking, Systems and Security (NSysS) (2017), 174–179.

Hossen, M. A., Siddika, F. & Chanda, T. K. A comparison of some soft computing methods on imbalanced data in International Conference on Cyber Security and Computer Science (2018).

Hossen, M. A. & Siddika, F. A Web Based Four-Tier Architecture using Reduced Feature Based Neural Network Approach for Prediction of Student Performance in 2021 2nd Int Conference on Robotics, Electrical and Signal Processing Techniques (ICREST) (2021), 269–273.

Hossen, M. A. & Siddika, F. Ensemble method based architecture using random forest importance to predict employee’s turn over in Journal of Physics: Conference Series 1755 (2021), 012039.

Hossen, M. A. & Siddika, F. Hybrid sampling and random forest based machine learning approach for software defect prediction in InECCE2019, Lecture Notes in Electrical Engineering) 632 (2019), 541–553.