Compositional Learning

Compositional learning aims to enable AI systems to understand and generate complex concepts by combining simpler, learned primitives, mirroring human cognitive abilities. Current research focuses on developing methods to effectively compose these primitives, employing various architectures including transformers, graph neural networks, and automata-based approaches, often within contrastive or adversarial learning frameworks. This research is significant because it addresses the limitations of current AI models in generalizing to unseen situations and complex tasks, paving the way for more robust and adaptable AI systems across diverse applications like image retrieval, natural language processing, and robotic control.