Learning as program induction
Neil R Bramley, Eric R. Schulz, Fei Xu, Joshua B. Tenenbaum · 2018
The notion that the mind approximates rational (Bayesian) inference has had a strong influence on thinking in psychology since the 1950s. In constrained scenarios, typical of psychology experiments, people often behave in ways that approximate the dictates of probability theory. However, natural learning contexts are typically much more open-ended --- there are often no clear limits on what is possible, and initial proposals often prove inadequate. This means that coming up with the right hypotheses and theories in the first place is often much harder than ruling among them. How do people, and how can machines, expand their hypothesis spaces to generate wholly new ideas, plans and solutions?Recent work has begun to shed light on this problem via the idea that many aspects of learning can be better understood through the mathematics of program induction.People are demonstrably able to compose hypotheses from parts and incrementally grow and adapt their models of the world. A number of recent studies has formalized these abilities as program induction, using algorithms that mix stochastic recombination of primitives with memoization and compression to explain data, ask informative questions [8], and support one- and few-shot-inferences. Program induction is also proving to be an important notion for understanding development and learning through play and the formation of geometric understanding about the physical world.The aim of this workshop is thus to bring together scientists who have a joint interest in how intelligent systems (humans or machines) can learn rich representations and action plans (expressable as programs) though observing and interacting with the world.