Playgol: learning programs through play
This addresses the challenge of developing program induction systems that can self-discover background knowledge, potentially benefiting AI and machine learning fields focused on automated reasoning and learning.
The paper tackles the problem of program induction by introducing an unsupervised playing stage where the system creates and solves its own tasks to discover reusable programs, which are then used in a supervised building stage to improve performance on user-supplied tasks. Experimental results in robot planning and string transformations suggest that playing can substantially improve learning performance, though no specific numerical gains are provided.
Children learn though play. We introduce the analogous idea of learning programs through play. In this approach, a program induction system (the learner) is given a set of tasks and initial background knowledge. Before solving the tasks, the learner enters an unsupervised playing stage where it creates its own tasks to solve, tries to solve them, and saves any solutions (programs) to the background knowledge. After the playing stage is finished, the learner enters the supervised building stage where it tries to solve the user-supplied tasks and can reuse solutions learnt whilst playing. The idea is that playing allows the learner to discover reusable general programs on its own which can then help solve the user-supplied tasks. We claim that playing can improve learning performance. We show that playing can reduce the textual complexity of target concepts which in turn reduces the sample complexity of a learner. We implement our idea in Playgol, a new inductive logic programming system. We experimentally test our claim on two domains: robot planning and real-world string transformations. Our experimental results suggest that playing can substantially improve learning performance. We think that the idea of playing (or, more verbosely, unsupervised bootstrapping for supervised program induction) is an important contribution to the problem of developing program induction approaches that self-discover BK.