Danial Safaei

Research interests

Safety assurance for AI-enabled autonomous systems; trustworthy evaluation and simulation-to-real transfer; explainable AI as a measurement instrument; synthetic-data fidelity; scenario-based testing.

Education

• 2024 — 2029 (exp.)

  PhD, Safe AI for Autonomous Systems

  WMG, University of Warwick, United Kingdom

  Supervisors: Prof. Siddartha Khastgir and Prof. Matthew Higgins (previously co-supervised by Dr Xingyu Zhao). Industrial partner: Siemens Digital Industries Software.

• 2019 — 2022

  MSc, Artificial Intelligence

  University of Tehran, Iran

  GPA 19.41 / 20.0.

• 2014 — 2019

  BSc, Computer Software Engineering

  University of Kashan, Iran

• 2010 — 2014

  High School Diploma, Mathematics and Physics

  National Organization for Development of Exceptional Talents (SAMPAD), Iran

Research and professional experience

• Feb 2025 — present

  Research Collaborator (PhD partnership)

  Siemens Digital Industries Software

  Industrial research partnership alongside the doctorate, linking simulation-fidelity research to virtual-validation practice.

• Oct 2024 — present

  PhD Researcher

  Safe Autonomy Research Group, WMG, University of Warwick

• Jan 2020 — Sep 2024

  AI Researcher / Reinforcement and Inverse Reinforcement Learning Researcher

  University of Tehran, Iran

  Deep learning, reinforcement and inverse reinforcement learning, computer vision, graph neural networks, and simulation.

• Jan 2021 — Sep 2024

  Foreign Exchange Trader (part-time)

  Windsor Brokers

Teaching and service

• Oct 2025 — present

  Teaching Assistant

  WMG, University of Warwick

• Jul 2025 — present

  PGR Academic Officer

  Postgraduate Society, University of Warwick

  Elected representative for postgraduate researchers on academic matters.

Peer-reviewed publications

1. 2026

   Quantifying Fidelity: A Decisive Feature Approach to Comparing Synthetic and Real Imagery

   D. Safaei, S. Khastgir, M. Alirezaei, J. Ploeg, C.-H. Cheng, S. Tong, X. Zhao

   2026 IEEE Intelligent Vehicles Symposium (IV), pp. 847–854. Detroit, MI, USA. DOI: 10.1109/IV66570.2026.11624013

2. 2024

   DeePLT: Personalized Lighting Facilitates by Trajectory Prediction of Recognized Residents in the Smart Home

   D. Safaei, A. Sobhani, A. A. Kiaei

   International Journal of Information Technology, vol. 16, no. 5, pp. 2987–2999. DOI: 10.1007/s41870-023-01665-1

Preprints and working papers

1. 2024

   A GraphSAGE Discovers Synergistic Combinations of Gefitinib, Paclitaxel, and Icotinib for Lung Adenocarcinoma Management: the RAIN Protocol

   S. Sadeghi, A. A. Kiaei, M. Boush, N. Salari, M. Mohammadi, D. Safaei, et al.

   medRxiv. DOI: 10.1101/2024.04.14.24304384

2. 2024

   Graph Attention Networks for Drug Combination Discovery: Targeting Pancreatic Cancer Genes with RAIN Protocol

   E. Parichehreh, A. A. Kiaei, M. Boush, D. Safaei, et al.

   medRxiv. DOI: 10.1101/2024.02.18.24302988

3. 2023

   Systematic Review and Network Meta-analysis of Drug Combinations Suggested by Machine Learning, with the Aim of Improving the Effectiveness of Ipilimumab in Treating Melanoma

   D. Safaei, A. A. Kiaei, M. Boush, S. Abadijou, et al.

   medRxiv. DOI: 10.1101/2023.05.13.23289940

4. 2023

   Drug Combinations Proposed by Machine Learning to Improve the Efficacy of Tecovirimat in the Treatment of Monkeypox

   M. Boush, A. A. Kiaei, D. Safaei, S. Abadijou, N. Salari, M. Mohammadi

   medRxiv. DOI: 10.1101/2023.04.23.23289008

5. 2023

   Recommending Drug Combinations Using Reinforcement Learning to Target Genes/Proteins that Cause Stroke

   M. Boush, A. A. Kiaei, D. Safaei, S. Abadijou, N. Salari, M. Mohammadi

   medRxiv. DOI: 10.1101/2023.04.20.23288906

6. 2023

   Recommending Drug Combinations Using Reinforcement Learning Targeting Genes/Proteins Associated with Heterozygous Familial Hypercholesterolemia

   A. A. Kiaei, M. Boush, S. Abadijou, S. Momeni, D. Safaei, et al.

   Research Square. DOI: 10.21203/rs.3.rs-2379891/v1

7. 2023

   Diagnosing Alzheimer's Disease Levels Using Machine Learning and MRI: A Novel Approach

   A. A. Kiaei, R. Bahadori, H. Malek Zadeh, M. Boush, S. Abadijou, D. Safaei, A. H. Mehdikhani

   Preprints.org. DOI: 10.20944/preprints202306.1184.v1

8. 2023

   Active Identity Function as Activation Function

   A. A. Kiaei, M. Boush, D. Safaei, S. Abadijou, et al.

   Preprints.org. DOI: 10.20944/preprints202305.1018.v1

9. 2023

   FPL: False Positive Loss

   A. A. Kiaei, M. Boush, D. Safaei, S. Abadijou, et al.

   Preprints.org. DOI: 10.20944/preprints202309.0280.v1

Bibliometrics

89 citations · h-index 6 · i10-index 5 · 11 works. Source: Google Scholar, August 2026.

Technical skills

Languages

Python, C++, MATLAB, R, SQL, Bash

Machine learning

PyTorch, TensorFlow, Keras, scikit-learn, Hugging Face, OpenCV

Autonomy and simulation

CARLA, SUMO, ROS, Gazebo, Unreal Engine, OpenSCENARIO

Methods

Explainable AI, counterfactual explanation, diffusion and generative models, graph neural networks, Bayesian and probabilistic modelling

Languages

Persian (native) · English (professional working proficiency)