About

1) Overview

Canadian AI graduate interested in the science and engineering of advanced intelligence and autonomy.

Research: I recently completed my second M.Sc. in Computer Science at the Mila AI Institute, where I explored state representation learning for deep reinforcement learning, advised by Pablo Samuel Castro. Our work together led to a survey of the field published in Transactions on Machine Learning Research. I also contributed to multi-agent reinforcement learning research at Mila, collaborating with Juan Duque under the supervision of Aaron Courville and Hugo Larochelle. Our work investigated opponent-shaping methods for improving cooperation in climate and economic multi-agent simulations, leading to a publication at ICLR 2026. Before joining Mila, I worked with Christian Ethier and his research group at the CERVO Brain Research Center, where I developed end-to-end pipelines for deep learning-based brain-computer interface decoding.

Industry: More recently, I applied AI in industry R&D at Bombardier Aerospace and Bentley Systems. At Bombardier, I developed flight-test data pipelines and customized neural network training pipelines for evaluating approaches to virtual sensor development. At Bentley Systems, I investigated different ways of using deep reinforcement learning for discrete optimization problems in civil engineering applications.

2) Research Interests

Autonomous Systems: Understanding how to perceive, represent, reason, plan, and act in complex environments. Topics include representation learning, world-model-based planning, uncertainty quantification, memory, control, and multi-agent coordination.

Toward Full Autonomy: Investigating what prevents current systems from operating reliably in dynamic, uncertain, and unfamiliar environments. Topics include decision-making under uncertainty, long-horizon and multi-agent planning, fault recovery, etc.

Science and Engineering: Advancing AI methods for scientific simulation, engineering design, and optimization, including surrogate and agentic approaches, differentiable simulators, physics-informed learning, and closing the simulation-to-reality gap.

3) Publications

• A Survey of State Representation Learning for Deep Reinforcement Learning

Transactions on Machine Learning Research (TMLR), 2025
Ayoub E., Pablo Samuel Castro Comprehensive survey of state representation learning methods, evaluation protocols, and open challenges in deep reinforcement learning.

• Towards Climate Investment Policies Informed by Reinforcement Learning

International Conference on Learning Representations, 2026
Juan Duque, Razvan C., Ayoub E., Aaron Courville, Hugo Larochelle
Study on opponent-shaping multi-agent reinforcement learning for cooperation in climate-economic social dilemmas.