Search Results for author:
Found 12 papers, 6 papers with code
Exact, Fast and Expressive Poisson Point Processes via Squared Neural Families
Maximum likelihood and maximum a posteriori estimates in a reparameterisation of the final layer of the intensity function can be obtained by solving a (strongly) convex optimisation problem using projected gradient descent.
Gaussian Ensemble Belief Propagation for Efficient Inference in High-Dimensional Systems
The use of local messages in a graphical model structure ensures that the approach is suited to distributed computing and can efficiently handle complex dependence structures.
Scalable Optimal Transport Methods in Machine Learning: A Contemporary Survey
This paper is about where and how optimal transport is used in machine learning with a focus on the question of scalable optimal transport.
Deep equilibrium models as estimators for continuous latent variables
We consider a generalised setting where the canonical parameters of the exponential family are a nonlinear transformation of the latents.
Efficient Gaussian Process Model on Class-Imbalanced Datasets for Generalized Zero-Shot Learning
Zero-Shot Learning (ZSL) models aim to classify object classes that are not seen during the training process.
Gaussian Process Bandits with Aggregated Feedback
We consider the continuum-armed bandits problem, under a novel setting of recommending the best arms within a fixed budget under aggregated feedback.
Declarative nets that are equilibrium models
We show that using a kernelised generalised linear model (kGLM) as an inner problem in a DDN yields a large class of commonly used DEQ architectures with a closed-form expression for the hidden layer parameters in terms of the kernel.
Avoiding Kernel Fixed Points: Computing with ELU and GELU Infinite Networks
Secondly, and more generally, we analyse the fixed-point dynamics of iterated kernels corresponding to a broad range of activation functions.
Richer priors for infinitely wide multi-layer perceptrons
The model resulting from partially exchangeable priors is a GP, with an additional level of inference in the sense that the prior and posterior predictive distributions require marginalisation over hyperparameters.
Expressive Priors in Bayesian Neural Networks: Kernel Combinations and Periodic Functions
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 kernel can capture seasonal variation with a long term trend.
Exchangeability and Kernel Invariance in Trained MLPs
In the analysis of machine learning models, it is often convenient to assume that the parameters are IID.
Invariance of Weight Distributions in Rectified MLPs
An interesting approach to analyzing neural networks that has received renewed attention is to examine the equivalent kernel of the neural network.
