Jared Dunnmon
Stanford University
59 Papers
303 Citations
Jared Dunnmon is an academic researcher from Stanford University. The author has contributed to research in topics: Computer science & Deep learning. The author has an hindex of 17, co-authored 48 publications. Previous affiliations of Jared Dunnmon include Duke University.
Chat about Author
Papers
Hidden stratification causes clinically meaningful failures in machine learning for medical imaging
Luke Oakden-Rayner,Jared Dunnmon,Gustavo Carneiro,Christopher Ré +3 more
- 02 Apr 2020
TL;DR: Evidence is found that hidden stratification can occur in unidentified imaging subsets with low prevalence, low label quality, subtle distinguishing features, or spurious correlates, and that it can result in relative performance differences of over 20% on clinically important subsets.
401
•Posted Content
Learning to Compose Domain-Specific Transformations for Data Augmentation
TL;DR: In this article, a generative adversarial approach is proposed to learn a sequence model over user-specified transformation functions using GANs, which can make use of arbitrary, non-deterministic transformation functions, and is robust to misspecified user input.
Training Complex Models with Multi-Task Weak Supervision.
Alexander Ratner,Braden Hancock,Jared Dunnmon,Frederic Sala,Shreyash Pandey,Christopher Ré +5 more
- 17 Jul 2019
TL;DR: This work shows that by solving a matrix completion-style problem, it can recover the accuracies of these multi-task sources given their dependency structure, but without any labeled data, leading to higher-quality supervision for training an end model.
Power extraction from aeroelastic limit cycle oscillations
TL;DR: In this paper, a flexible beam with piezoelectric laminates is excited by a uniform axial flow field in a manner analogous to a flapping flag such that the system delivers power to an electrical impedance load.
247
•Proceedings Article
Learning to Compose Domain-Specific Transformations for Data Augmentation.
Alexander Ratner,Henry R. Ehrenberg,Zeshan Hussain,Jared Dunnmon,Christopher Ré +4 more
- 01 Dec 2017
TL;DR: In this article, a generative adversarial approach is proposed to learn a sequence model over user-specified transformation functions using GANs, which can make use of arbitrary, non-deterministic transformation functions, and is robust to misspecified user input.