ajwieme at gmail dot com

I am finishing my PhD this Fall and on the job market!

I am a fourth year PhD student in Computer Science at The University of Colorado, Boulder, working under the supervision of Katharina von der Wense in the NALA lab. Previously, I worked on NLP research as a research scientist at Pearson, and before that I was a researcher at the US Army Research laboratory. I have also worked as a software engineer at several companies.

Research Interests

My PhD has focused on NLP systems geared towards underrepresented languages and computational morphology. This often entails using unsupervised and semi-supervised machine learning methods, and training deep learning models on very sparse datasets. Currently, I am focused on two things:
  • Using reinforcement learning to align unsupervised training data with particular supervised tasks
  • How better tokenization can enable multilingual LMs to perform well for underrepresented languages.

I additionally maintain open-source research software (Yoyodyne) for deep learning experiments geared towards the computational morphology and phonology communities with Kyle Gorman and Travis Bartley.

News

Sep. 2024 New paper accepted to EMNLP 2024! Getting The Most Out of Your Training Data: Exploring Unsupervised Tasks for Morphological Inflection. This was a masters project that I designed and mentored.
Aug. 2024 Submitted a demo paper for our software Yoyodyne
May. 2024 Started a summer machine learning scientist internship at Apple 
  • Focused on detecting cognitive decline from speech.
Jan. 2024 Our Paper "Quantifying the Hyperparameter Sensitivity of Neural Networks for Character-level Sequence-to-Sequence Tasks" was accepted to EACL 2024!
Dec. 2023 I am a PhD candidate---I passed my candidacy exam and have proposed my dissertation!
May. 2023 Our Paper "An Investigation of Noise in Morphological Inflection" was accepted to Findings of ACL 2023.
Oct. 2022 New Paper "A Comprehensive Comparison of Neural Networks as Cognitive Models of Inflection" was accepted to EMNLP 2022!
Apr. 2022 This summer I will be interning in the content generation group at ETS AI Labs.
  • Focused on RAG for argument generation with fact sand events.
Mar. 2022 Our paper "Morphological Processing of Low-Resource Languages: Where We Are and What's Next" was accepted to Findings of ACL 2022! See the preprint here
Aug. 2021 The shared task has completed. We have published a paper on the results: "Findings of the SIGMORPHON 2021 Shared Task on Unsupervised Morphological Paradigm Clustering".
Mar. 2021 Were organizing a shared task on Unsupervised Morphological Paradigm Clustering. Register to participate here!
Jan. 2021 Starting PhD in Natural Language Processing