A novel method for sampling a class of subsequential string transductions encoding homomorphisms allows rigorous testing of learning models' capacity for compositionality.
My doctoral dissertation, examining the relationship between abductive inference and algebraically structured hypothesis spaces, giving a general form for grammar learning over arbitrary linguistic structure.
We derive the well-studied subregular classes of formal languages, which computationally characterize natural language typology, purely from the perspective of algorithmic learning problems.
We formalize various iterative prosodic processes including stress, syllabification and epenthesis using logical graph transductions, showing that the necessary use of fixed point operators without quantification restricts them to a structured subclass of subsequential functions.