activity
20182023
most citedOn Uncertainty, Tempering, and Data Augmentation in Bayesian Classification

9 citations · 26 across the 5 of their papers we have counts for

collaborators

7 papers

cs.LG2023

Should We Learn Most Likely Functions or Parameters?

Shikai Qiu, Tim G. J. Rudner, Sanyam Kapoor +1

Standard regularized training procedures correspond to maximizing a posterior distribution over parameters, known as maximum a posteriori (MAP) estimation. However, model parameter…

cs.LG20227 cited

PAC-Bayes Compression Bounds So Tight That They Can Explain Generalization

Sanae Lotfi, Marc Finzi, Sanyam Kapoor +3

While there has been progress in developing non-vacuous generalization bounds for deep neural networks, these bounds tend to be uninformative about why deep learning works. In this…

cs.LG20228 cited

Pre-Train Your Loss: Easy Bayesian Transfer Learning with Informative Priors

Ravid Shwartz-Ziv, Micah Goldblum, Hossein Souri +4

Deep learning is increasingly moving towards a transfer learning paradigm whereby large foundation models are fine-tuned on downstream tasks, starting from an initialization learne…

cs.LG20229 cited

On Uncertainty, Tempering, and Data Augmentation in Bayesian Classification

Sanyam Kapoor, Wesley J. Maddox, Pavel Izmailov +1

Aleatoric uncertainty captures the inherent randomness of the data, such as measurement noise. In Bayesian regression, we often use a Gaussian observation model, where we control t…

cs.LG20212 cited

SKIing on Simplices: Kernel Interpolation on the Permutohedral Lattice for Scalable Gaussian Processes

Sanyam Kapoor, Marc Finzi, Ke Alexander Wang +1

State-of-the-art methods for scalable Gaussian processes use iterative algorithms, requiring fast matrix vector multiplies (MVMs) with the covariance kernel. The Structured Kernel…

cs.LG2019

First-Order Preconditioning via Hypergradient Descent

Ted Moskovitz, Rui Wang, Janice Lan +4

Standard gradient descent methods are susceptible to a range of issues that can impede training, such as high correlations and different scaling in parameter space.These difficulti…