Jakob Nicolaus Foerster

multi-agent reinforcement learning (1) reinforcement learning (1) function approximation (1) attention mechanism (1) policy learning (1) model merging (1) adversarial attack (1) inductive bia (1) language model (1) game ai (1) cooperative intelligence (1) mixed-motive game (1) social generalization (1) expert specialization (1) continual pretraining (1) multi-agent coordination (1) intuitive policy (1) attention-based architecture (1) action feature (1) interpretable policy (1)