Amy X. Lu
University of Toronto
12 Papers
13 Citations
Amy X. Lu is an academic researcher from University of Toronto. The author has contributed to research in topics: Computer science & Covariate. The author has an hindex of 5, co-authored 11 publications. Previous affiliations of Amy X. Lu include Ontario Institute for Cancer Research.
Chat about Author
Papers
Hurtful words: quantifying biases in clinical contextual word embeddings
Haoran Zhang,Amy X. Lu,Mohamed Abdalla,Matthew B. A. McDermott,Marzyeh Ghassemi +4 more
- 02 Apr 2020
TL;DR: The authors examined the extent to which embeddings may encode marginalized populations differently, and how this may lead to a perpetuation of biases and worsened performance on clinical tasks, and found that classifiers trained from BERT representations exhibit statistically significant differences in performance, often favoring the majority group with regards to gender, language, ethnicity, and insurance status.
106
History and publication trends in the diffusion and early uptake of indirect comparison meta-analytic methods to study drugs: animated coauthorship networks over time.
TL;DR: Although Canada and the USA were the first to apply indirect comparison meta-analytic methods, Europe led their diffusion and the increase in uptake of these methods may have been facilitated by acceptance by regulatory agencies, which are calling for more comparative drug effect data to assist in drug accessibility and reimbursement decisions.
13
Comparative toxicoproteogenomics of mouse and rat liver identifies TCDD-resistance genes
Stephenie D. Prokopec,Amy X. Lu,Amy X. Lu,Sandy Che Eun S. Lee,Cindy Q. Yao,Ren X. Sun,John D. Watson,Rabah Soliymani,Richard de Borja,Ada Wong,Michelle Sam,Philip C. Zuzarte,John Douglas Mcpherson,Allan B. Okey,Raimo Pohjanvirta,Raimo Pohjanvirta,Paul C. Boutros +16 more
TL;DR: The value of integrating genomic, transcriptomic and proteomic evidence across multi-species models in toxicologic studies is shown, with several hepatic proteins showed parallel up- or downward alterations with their RNAs, with three genes showing consistent, strain-dependent changes.
Pretraining strategies for effective promoter-driven gene expression prediction
Aniketh Janardhan Reddy,Michael Howland Herschl,Sathvik Kolli,Amy X. Lu,Xinyang Geng,Aviral Kumar,Patrick D. Hsu,Sergey Levine,Nilah M. Ioannidis +8 more
TL;DR: In this paper , the authors proposed a method for modeling cell type-specific expression from compact promoters and demonstrated the effectiveness of pretraining on existing promoter-driven expression datasets from other cell types.
•Posted Content
The Cells Out of Sample (COOS) dataset and benchmarks for measuring out-of-sample generalization of image classifiers
TL;DR: The COOS-7 (Cells Out Of Sample 7-Class) dataset as discussed by the authors was created to measure the generalization capacity of image classifiers, as we can image the same classes of objects under increasingly divergent, but controlled factors of variation.
6