CV
Research, education, publications, open-source software, and professional experience.
Contact Information
| Name | Amine Andam |
| Professional Title | PhD Candidate in Reinforcement Learning |
| andamamine83@gmail.com | |
| Phone | +212 652 865 020 |
| Website | https://github.com/AmineAndam04 |
Professional Summary
PhD candidate in Reinforcement Learning. I focus on both robust RL and adversarial attacks on RL agents. I also develop open-source libraries for RL and multi-agent RL in PyTorch and JAX.
Education
-
2022 - present Benguerir, Morocco
PhD
Mohammed VI Polytechnic University
Adversarial Reinforcement Learning
-
2018 - 2021 Rabat, Morocco
Engineering Degree
National Institute of Statistics and Applied Economics
Quantitative Finance
Publications
-
2026 Robust Policy Optimization via Adversarial Importance Sampling
Preprint
A. Andam, J. Bentahar, and M. Hedabou.
-
2026 A Unified and Constrained View of Regularization-Based Robust Reinforcement Learning
Preprint
A. Andam, J. Bentahar, and M. Hedabou.
-
2025 Constrained Black-Box Attacks Against Cooperative Multi-Agent Reinforcement Learning
Preprint
A. Andam, J. Bentahar, and M. Hedabou.
-
2024 Multimodal Deep Reinforcement Learning for Visual Security of Virtual Reality Applications
IEEE Internet of Things Journal, vol. 11, no. 24
A. Andam, J. Bentahar, and M. Hedabou.
-
2026 Decentralized Security for Multi-User VR Applications Using Multi-Agent Reinforcement Learning
Internet of Things
A. Andam, J. Bentahar, E. Amhoud, M. Younas, and M. Hedabou.
-
2026 Scalable and Efficient Deep Reinforcement Learning-Based Model Checker for Computation Tree Logic
IEEE Transactions on Neural Networks and Learning Systems
G. Alwhishi, J. Bentahar, A. Andam, A. Elwhishi, and M. Hedabou.
-
2026 Securing Vehicular Ad Hoc Networks via Digital Twin-Driven Anomaly Detection and Graph Reinforcement Learning
IEEE Transactions on Vehicular Technology
L. El Bouga, A. Andam, J. Bentahar, E. Amhoud, and M. Hedabou.
Open Source Projects
-
advrl
A modular PyTorch library for training and evaluating robust Reinforcement Learning agents.
-
cleanmarl
A PyTorch library providing clean, single-file implementations of multi-agent Reinforcement Learning algorithms, along with detailed documentation and educational resources.
-
marlbench
A library that provides a unified API, common wrappers, and parallel execution for multi-agent Reinforcement Learning environments.
-
flaxrl
Fast implementations of single-agent and multi-agent Reinforcement Learning algorithms using JAX and Flax NNX.
Experience
-
2021 - 2021 Casablanca, Morocco
Data Science Intern
Mazars
- March–July 2021: Worked on Machine Learning applications in insurance.
-
2020 - 2020 Casablanca, Morocco
Data Science Intern
RMA Capital
- July–September 2020: Developed a Python library for algorithmic trading.