broadly speaking, i’m interested in machine learning and applying it to language.
here are some relevant research themes and associated questions:
lower-resourced + morphologically rich languages
❓ how can we learn representations that systematically capture morphology?
❓ can we inform cross-lingual transfer learning by linguistic relatedness?
representation learning & structured prediction
❓ can we replace/augment vectors with structured representations like lattices or graphs?
❓ how can we incorporate expert information into our representations in a controllable way?
bayesian methods and model comparison
❓ how do we really know that model a performed better than model b on some task/dataset?
❓ how can we add interpretable uncertainty estimates to our model evaluation?
❓ what is the population we are trying to generalize to when using e.g. shared tasks?
for an up-to-date list of publications, see my google scholar profile.