Neural diffusion processes
V Dutordoir, A Saul, Z Ghahramani… - … on Machine Learning, 2023 - proceedings.mlr.press
Neural network approaches for meta-learning distributions over functions have desirable
properties such as increased flexibility and a reduced complexity of inference. Building on …
properties such as increased flexibility and a reduced complexity of inference. Building on …
Episodic multi-task learning with heterogeneous neural processes
This paper focuses on the data-insufficiency problem in multi-task learning within an
episodic training setup. Specifically, we explore the potential of heterogeneous information …
episodic training setup. Specifically, we explore the potential of heterogeneous information …
Geometric neural diffusion processes
Denoising diffusion models have proven to be a flexible and effective paradigm for
generative modelling. Their recent extension to infinite dimensional Euclidean spaces has …
generative modelling. Their recent extension to infinite dimensional Euclidean spaces has …
The neural process family: Survey, applications and perspectives
The standard approaches to neural network implementation yield powerful function
approximation capabilities but are limited in their abilities to learn meta representations and …
approximation capabilities but are limited in their abilities to learn meta representations and …
Practical equivariances via relational conditional neural processes
Abstract Conditional Neural Processes (CNPs) are a class of metalearning models popular
for combining the runtime efficiency of amortized inference with reliable uncertainty …
for combining the runtime efficiency of amortized inference with reliable uncertainty …
Spatial multi-attention conditional neural processes
LL Bao, JS Zhang, CX Zhang - Neural Networks, 2024 - Elsevier
Spatial prediction tasks are challenging when observed samples are sparse and prediction
samples are abundant. Gaussian processes (GPs) are commonly used in spatial prediction …
samples are abundant. Gaussian processes (GPs) are commonly used in spatial prediction …
Environmental sensor placement with convolutional Gaussian neural processes
Environmental sensors are crucial for monitoring weather conditions and the impacts of
climate change. However, it is challenging to place sensors in a way that maximises the …
climate change. However, it is challenging to place sensors in a way that maximises the …
Diffusion-Augmented Neural Processes
L Bonito, J Requeima, A Shysheya… - arXiv preprint arXiv …, 2023 - arxiv.org
Over the last few years, Neural Processes have become a useful modelling tool in many
application areas, such as healthcare and climate sciences, in which data are scarce and …
application areas, such as healthcare and climate sciences, in which data are scarce and …
Neural processes with event triggers for fast adaptation to changes
P Brunzema, P Kruse, S Trimpe - 6th Annual Learning for …, 2024 - proceedings.mlr.press
Traditionally, first-principle models are used to monitor and control dynamical systems.
However, modeling complex systems using first principles can be challenging. Learning the …
However, modeling complex systems using first principles can be challenging. Learning the …
Memory efficient neural processes via constant memory attention block
Neural Processes (NPs) are popular meta-learning methods for efficiently modelling
predictive uncertainty. Recent state-of-the-art methods, however, leverage expensive …
predictive uncertainty. Recent state-of-the-art methods, however, leverage expensive …