Amos Storkey

Object-Centric World Models from Few-Shot Annotations for Sample-Efficient Reinforcement Learning

Meta-Adaptive Prompt Distillation for Few-Shot Visual Question Answering

Forgetting is Everywhere

HyperMARL: Adaptive Hypernetworks for Multi-Agent RL

Objects matter: object-centric world models improve reinforcement learning in visually complex environments

EEVEE and GATE: Finding the right benchmarks and how to run them seamlessly

Towards Realistic Hyperparameter Optimization in Continual Learning