repositories

A selection of research and teaching repositories. For a full list, visit my GitHub profile.

Research

alem-env alem-world

JAX-based benchmark for open-ended multi-agent coordination in language agents. Under review.

Python — LLM Agents MARL JAX Benchmark
probing-dec-pomdps

Information-theoretic probes for testing whether MARL benchmarks require genuine decentralised reasoning. AAMAS 2026 Oral.

Python — MARL Probing JAX
HyperMARL

Adaptive hypernetworks for multi-agent RL. NeurIPS 2025.

Python — MARL Hypernetworks JAX
Mava instadeepai

Research framework for distributed multi-agent reinforcement learning.

Python — MARL Distributed RL JAX
DQN-Atari

DQN implementation for Atari benchmarks.

Python — Deep RL Atari PyTorch

Teaching

indaba-pracs-2022 deep-learning-indaba

Practical sessions for the Deep Learning Indaba 2022.

Jupyter Notebook — Machine Learning Deep Learning Reinforcement Learning
indaba-pracs-2023 deep-learning-indaba

Practical sessions for the Deep Learning Indaba 2023.

Jupyter Notebook — Machine Learning Deep Learning Reinforcement Learning
indaba-pracs-2024 deep-learning-indaba

Practical sessions for the Deep Learning Indaba 2024.

Jupyter Notebook — Machine Learning Deep Learning Reinforcement Learning