Lifelong Learning of Compositional Structures
This addresses the problem of enabling AI systems to reuse knowledge across related tasks, which is incremental as it combines existing lines of work.
The paper tackles the challenge of learning compositional structures in artificial systems by integrating lifelong learning with compositional learning, presenting a framework that separates learning into combining existing components and adapting them to new tasks, and empirically demonstrates handling the stability-flexibility trade-off.
A hallmark of human intelligence is the ability to construct self-contained chunks of knowledge and adequately reuse them in novel combinations for solving different yet structurally related problems. Learning such compositional structures has been a significant challenge for artificial systems, due to the combinatorial nature of the underlying search problem. To date, research into compositional learning has largely proceeded separately from work on lifelong or continual learning. We integrate these two lines of work to present a general-purpose framework for lifelong learning of compositional structures that can be used for solving a stream of related tasks. Our framework separates the learning process into two broad stages: learning how to best combine existing components in order to assimilate a novel problem, and learning how to adapt the set of existing components to accommodate the new problem. This separation explicitly handles the trade-off between the stability required to remember how to solve earlier tasks and the flexibility required to solve new tasks, as we show empirically in an extensive evaluation.