DR Daniel Rashedi

PhD student · TU Hamburg

Daniel Rashedi

PhD Student

Institute for Software Systems
Hamburg University of Technology

01

What I explore

Research interests

01

Fault Localization for Neural Networks

Spectrum-based techniques adapted for neural network debugging and analysis.

02

Scalability & Efficiency

Making software engineering methods for neural networks practical at scale.

02

All publications

Publications

4 publications
01 Conference paper 2026

Automatically Generating Programming Exercises with Open-Source LLMs: Integrating Lecture Slides and Learning Objectives

M. B. Hoffmann, D. Rashedi, and S. Schupp

15th International Workshop on Trends in Functional Programming in Education (TFPiE 2026), Odense, Denmark.

02 Conference paper 2025

Efficient Hit-Spectrum-Guided Fast Gradient Sign Method: An Adjustable Approach with Memory and Runtime Optimizations

D. Rashedi and S. Schupp

20th International Conference on Software Technologies (ICSOFT 2025), Bilbao, Spain.

✦ Nominated for Best Paper Award
doipdf
03 Conference paper 2024

Repairing Neural Networks for Image Classification Problems Using Spectrum-Based Fault Localization

D. Rashedi and S. Schupp

4th IEEE International Conference on Software Engineering and Artificial Intelligence (SEAI 2024), Xiamen, China.

doi
04 Journal article 2025

Efficient Adversarial Generation through Selective SBFL Guidance and Automated Sub-Model Strategy

D. Rashedi and S. Schupp

Communications in Computer and Information Science (CCIS).