My research sits at the intersection of causality and reinforcement learning — understanding how causal reasoning can make RL agents more robust, efficient, and interpretable. Find my full profile on Google Scholar.
- 2026Submitted to SAB 2026. An active-inference agent allocating a fixed observation-precision budget toward whichever bodily channel its own posterior flags as most needed more than doubles learning-phase survival vs uniform precision; reversing the direction of the selector falls below uniform. The shaped likelihood reaches both belief update and the EFE planner, with the propagation to the planner carrying most of the gain.
- 2023Introduces an algorithm for active learning during model-based planning, combining active inference with model-based RL to improve sample efficiency and uncertainty-aware decision-making.
- 2024MSc thesis surveying the current landscape of causal methods applied to reinforcement learning — covering key open problems and promising research directions.
- 2021Reviews the state of causal reasoning in multi-agent settings and identifies the key open problems at the frontier of causal MARL research.
- 2021Introduces Mava, a set of MARL components and system abstractions built on top of dm-acme for scalable, distributed training across hardware accelerators. Work done at InstaDeep. I contributed to the original system design and implemented the QMIX algorithm. A revised version of the paper and framework was released in 2023.
- 2021Surveys counterfactual reasoning methods through the lens of Pearl's causal hierarchy, focusing on their application to sequential decision-making and RL.
- 2021Surveys model-based RL approaches that use learned world models for planning, exploration, and sample-efficient learning in complex environments.