E
Ethan Caballero
Publications - 7
Citations - 490
Ethan Caballero is an academic researcher. The author has contributed to research in topics: Generalization & Robustness (computer science). The author has an hindex of 4, co-authored 6 publications receiving 152 citations.
Papers
More filters
Posted Content
Out-of-Distribution Generalization via Risk Extrapolation (REx)
David Krueger,Ethan Caballero,Joern-Henrik Jacobsen,Amy Zhang,Jonathan Binas,Dinghuai Zhang,Rémi Le Priol,Aaron Courville +7 more
TL;DR: This work introduces the principle of Risk Extrapolation (REx), and shows conceptually how this principle enables extrapolation, and demonstrates the effectiveness and scalability of instantiations of REx on various OoD generalization tasks.
Posted Content
In Search of Robust Measures of Generalization.
Gintare Karolina Dziugaite,Alexandre Drouin,Brady Neal,Nitarshan Rajkumar,Ethan Caballero,Linbo Wang,Ioannis Mitliagkas,Daniel M. Roy +7 more
TL;DR: This work addresses the question of how to evaluate generalization bounds empirically and argues that generalization measures should instead be evaluated within the framework of distributional robustness.
Proceedings Article
In search of robust measures of generalization
Gintare Karolina Dziugaite,Alexandre Drouin,Brady Neal,Nitarshan Rajkumar,Ethan Caballero,Linbo Wang,Ioannis Mitliagkas,Daniel M. Roy +7 more
TL;DR: The authors argue that generalization measures should instead be evaluated within the framework of distributional robustness, i.e., why the particular way the community now trains networks to achieve small training error also leads to small error on held-out data from the same population.
Proceedings ArticleDOI
Broken Neural Scaling Laws
TL;DR: A smoothly broken power law functional form that accurately models and extrapolates the scaling behaviors of deep neural networks for each task within a large and diverse set of upstream and downstream tasks, in zero-shot, prompted, and fine-tuned settings.
Proceedings Article
Out-of-Distribution Generalization via Risk Extrapolation (REx)
David Krueger,Ethan Caballero,Joern-Henrik Jacobsen,Amy Zhang,Jonathan Binas,Dinghuai Zhang,Rémi Le Priol,Aaron Courville +7 more
TL;DR: In this paper, risk extrapolation is used as a form of robust optimization over a perturbation set of extrapolated domains, and a penalty on the variance of training risks (V-REx) is proposed as a simpler variant.