Self-evolving scientific agents capable of conquering the hard tail of formal mathematics require Compositional Learning Behaviours (CLBs) -- the capacity to ground and recombine novel symbolic structures in context, beyond mere recombination of …
Suite of OpenAI Gym-compatible multi-agent reinforcement learning environment to benchmark for behavioral traits pertaining to symbolic behaviours (primarily: being receptive, constructive, malleable, and separable) using referential games.
Human beings use compositionality to generalise from past experiences to novel experiences. We assume a separation of our experiences into fundamental atomic components that can be recombined in novel ways to support our ability to engage with novel …