Timothy J. O'Donnell

Learn More
We present a model of unsupervised phono-logical lexicon discovery—the problem of simultaneously learning phoneme-like and word-like units from acoustic input. Our model builds on earlier models of unsuper-vised phone-like unit discovery from acoustic data (Lee and Glass, 2012), and unsuper-vised symbolic lexicon discovery using the Adaptor Grammar(More)
Parallel computers have matured to the point where they are capable of running a significant production workload. Characterizing this workload, however, is far more complicated than for the single-processor case. Besides the varying number of processors that may be invoked, the nodes themselves may provide differing computational resources (memory size, for(More)
In morphology, researchers have provided compelling evidence for the storage of even fully compositional structures that could otherwise be computed by rule. For example, a high-frequency word composed of multiple morphemes (e.g., root + plural inflection) may be stored directly rather than computed on the fly (e.g., Baayen, Dijkstra, & Schreuder, 1997).(More)
We present a model for inducing sen-tential argument structure, which distinguishes arguments from optional modi-fiers. We use this model to study whether representing an argument/modifier distinction helps in learning argument structure, and whether a linguistically-natural argu-ment/modifier distinction can be induced from distributional data alone. Our(More)
  • 1