activity
20182020
most citedExpressive Priors in Bayesian Neural Networks: Kernel Combinations and Periodic Functions

13 citations · 16 across the 2 of their papers we have counts for

collaborators

6 papers

cs.CV20203 cited

Structured Weight Priors for Convolutional Neural Networks

Tim Pearce, Andrew Y. K. Foong, Alexandra Brintrup

Selection of an architectural prior well suited to a task (e.g. convolutions for image data) is crucial to the success of deep neural networks (NNs). Conversely, the weight priors…

cs.LG2020

Avoiding Kernel Fixed Points: Computing with ELU and GELU Infinite Networks

Russell Tsuchida, Tim Pearce, Chris van der Heide +2

Analysing and computing with Gaussian processes arising from infinitely wide neural networks has recently seen a resurgence in popularity. Despite this, many explicit covariance fu…

stat.ML201913 cited

Expressive Priors in Bayesian Neural Networks: Kernel Combinations and Periodic Functions

Tim Pearce, Russell Tsuchida, Mohamed Zaki +2

A simple, flexible approach to creating expressive priors in Gaussian process (GP) models makes new kernels from a combination of basic kernels, e.g. summing a periodic and linear…

cs.LG2018

Bayesian Neural Network Ensembles

Tim Pearce, Mohamed Zaki, Andy Neely

Ensembles of neural networks (NNs) have long been used to estimate predictive uncertainty; a small number of NNs are trained from different initialisations and sometimes on differi…

stat.ML2018

Uncertainty in Neural Networks: Approximately Bayesian Ensembling

Tim Pearce, Felix Leibfried, Alexandra Brintrup +2

Understanding the uncertainty of a neural network's (NN) predictions is essential for many purposes. The Bayesian framework provides a principled approach to this, however applying…

stat.ML2018

Bayesian Inference with Anchored Ensembles of Neural Networks, and Application to Exploration in Reinforcement Learning

Tim Pearce, Nicolas Anastassacos, Mohamed Zaki +1

The use of ensembles of neural networks (NNs) for the quantification of predictive uncertainty is widespread. However, the current justification is intuitive rather than analytical…